Mtech AI Syllabus

SEMESTER I
Discipline: INTERDISCIPLINARY
Stream: ID2 (ARTIFICIAL INTELLIGENCE)
CATEGORY L T P CREDIT
221TCS009 MACHINE LEARNING 3 0 0 3
DISCIPLINE
CORE 1
Preamble: This course introduces machine learning concepts and popular machine learning
algorithms. It will cover the standard and most popular supervised learning algorithms
including linear regression, logistic regression, decision trees, k-nearest neighbour, an
introduction to Bayesian learning and the naive Bayes algorithm, support vector machines and
kernels and basic clustering algorithms. Dimensionality reduction methods and some
applications to real world problems will also be discussed. It helps the learners to develop
application machine learning based solutions for real world applications.
Course Outcomes:
After the completion of the course the student will be able to:*
CO 1 Analyse the Machine Learning concepts, classifications of Machine Learning

algorithms and basic parameter estimation methods. (Cognitive Knowledge Level:
Analyse)
CO 2 Illustrate the concepts of regression and classification techniques (Cognitive
Knowledge Level: Apply)
CO 3 Describe unsupervised learning concepts and dimensionality reduction techniques.
(Cognitive Knowledge Level: Apply)
CO 4 Explain Support Vector Machine concepts and graphical models. (Cognitive
CO 5 Choose suitable model parameters for different machine learning techniques and to
evaluate a model performance.(Cognitive Knowledge Level: Apply)
CO6 Design, implement and analyse machine learning solution for a real world problem.
(Cognitive Knowledge Level: Create)
Program Outcomes (PO)

Outcomes are the attributes that are to be demonstrated by a graduate after completing the
course.
PO1: An ability to independently carry out research/investigation and developmentwork in

engineering and allied streams
PO2: An ability to communicate effectively, write and present technical reports on complex
engineering activities by interacting with the engineering fraternity and with society at large.
PO3: An ability to demonstrate a degree of mastery over the area as per the specialization of the
program. The mastery should be at a level higher than the requirements in the appropriate
bachelor program
PO4: An ability to apply stream knowledge to design or develop solutions for real world problems by
following the standards
PO5: An ability to identify, select and apply appropriate techniques, resources and state-of-the-art
tool to model, analyse and solve practical engineering problems.
PO6: An ability to engage in life-long learning for the design and development related to the stream
related problems taking into consideration sustainability, societal, ethical and environmental
aspects
PO7: An ability to develop cognitive load management skills related to project management and
finance which focus on Entrepreneurship and Industry relevance.
Mapping of course outcomes with program outcomes
PO 1 PO 2 PO 3 PO 4 PO 5 PO 6 PO 7
CO 1
CO 2
CO 3
CO 4
CO 5
CO 6
Assessment Pattern
Bloom’s Category End Semester Examination
Apply 60-80%
Analyse 20-40%
Evaluate
Create
Mark distribution
Total CIE ESE ESE

Marks Duration
100 40 60 2.5 hours
Continuous Internal Evaluation Pattern:

Evaluation shall only be based on application, analysis or design based questions (for both
internal and end semester examinations).
Continuous Internal Evaluation : 40 marks

Micro project/Course based project : 20 marks
Course based task/Seminar/Quiz : 10 marks
Test paper, 1 no. : 10 marks
The project shall be done individually. Group projects not permitted.
Test paper shall include minimum 80% of the syllabus.
Course based task/test paper questions shall be useful in the testing of knowledge, skills,
comprehension, application, analysis, synthesis, evaluation and understanding of the
students.
End Semester Examination Pattern:
The end semester examination will be conducted by the University. There will be two parts;
Part A and Part B. Part A contain 5 numerical questions with 1 question from each module,
having 5 marks for each question. (such questions shall be useful in the testing of knowledge,
skills, comprehension, application, analysis, synthesis, evaluation and understanding of the
students). Students shall answer all questions.
Part B will contain 7 questions (such questions shall be useful in the testing of overall
achievement and maturity of the students in a course, through long answer questions relating to
theoretical/practical knowledge, derivations, problem solving and quantitative evaluation), with
minimum one question from each module of which student should answer any five. Each
question can carry 7 marks.
Total duration of the examination will be 150 minutes.
Course Level Assessment Questions

Course Outcome 1 (CO1):
1. Suppose that X is a discrete random variable with the following probability mass
function: where 0 ≤ θ ≤ 1 is a parameter. The following 10 independent observations were
taken from such a distribution: (3, 0, 2, 1, 3, 2, 1, 0, 2, 1). What is the maximum likelihood
estimate of θ.
2. What is the difference between Maximum Likelihood estimation (MLE) and Maximum a
Posteriori (MAP) estimation?
3. A gamma distribution with parameters α, β has the following density function, where Γ(t)
is the gamma function.
If the posterior distribution is in the same family as the prior distribution, then we say that the
prior distribution is the conjugate prior for the likelihood function. Using the Gamma
distribution as a prior, show that the Exponential distribution is a conjugate prior of the
Gamma distribution. Also, find the maximum a posteriori estimator for the parameter of the
Exponential distribution as a function of α and β.
Course Outcome 2 (CO2)

1. How can we interpret the output of a two-class logistic regression classifier as a robability?
2. Calculate the output of the following neuron Y if the activation function is a binary
sigmoid.
3. Suppose you have a 3-dimensional input x = (x1, x2, x3) = (2, 2, 1) fully connected with
weights (0.5, 0.3, 0.2) to one neuron which is in the hidden layer with sigmoid activation
function. Calculate the output of the hidden layer neuron.
4. Consider the case of the XOR function in which the two points {(0, 0),(1, 1)} belong to
one class, and the other two points {(1, 0),(0, 1)} belong to the other class. Design a
multilayer perceptron for this binary classification problem.
5. Why does a single perceptron cannot simulate simple XOR function? Explain how this
limitation is overcome?
6. Consider a naive Bayes classifier with 3 boolean input variables, X1, X2 and X3, and one
boolean output, Y. How many parameters must be estimated to train such a naive Bayes
classifier? How many parameters would have to be estimated to learn the above classifier
if we do not make the naive Bayes conditional independence assumption?
Course Outcome 3(CO3):
1. Describe the basic operation of k-means clustering.
2. A Poisson distribution is used to model data that consists of non-negative integers.

Suppose you observe m integers in your training set. Your model assumption is that each
integer is sampled from one of two different Gaussian distributions. You would like to
learn this model using the EM algorithm. List all the parameters of the model. Derive the
E-step and M-step for this model.
3. A uni-variate Gaussian distribution is used to model data that consists of non-negative

integers. Suppose you observe m integers in your training set. Your model assumption is
that each integer is sampled from one of two different Gaussian distributions. You would
like to learn this model using the EM algorithm. List all the parameters of the model.
Derive the E-step and M-step for the model.
4. Suppose you want to cluster the eight points shown below using k-means
Assume that k = 3 and that initially the points are assigned to clusters as follows:
C1 = {x1, x2, x3}, C2 = {x4, x5, x6}, C3 = {x7, x8}. Apply the k-means algorithm
until convergence, using the Manhattan distance.
1. Describe how Support Vector Machines can be extended to make use of kernels. Illustrate
with reference to the Gaussian kernel K(x, y) = e−y, where y = (x−y)2 .
2. Suppose that you have a linear support vector machine(SVM) binary classifier. Consider a
point that is currently classified correctly, and is far away from the decision boundary. If
you remove the point from the training set, and re-train the classifier, will the decision
boundary change or stay the same? Justify your answer.
3. What is the primary motivation for using the kernel trick in machine learning algorithms?
4. Show that the Boolean function (x1 ∧ x2) ∨ (¬x1 ∧ ¬x2) is not linearly separable (i.e. there
is no linear classifier sign(w1 x1 + w2 x2 + b) that classifies all 4 possible input points
correctly). Assume that “true” is represented by 1 and “false” is represented by −1. Show
that there is a linear separator for this Boolean function when we use the kernel K(x, y) =
(x · y)2 (x.y denotes the ordinary inner product) . Give the weights and the value of b for
one such separator.
5. Consider the following one dimensional training data set, ’x’ denotes negative examples
and ’o’ positive examples. The exact data points and their labels are given in the table.
Suppose a SVM is used to classify this data. Indicate which are the support vectors and
mark the decision boundary. Give the value of the cost function and of the model
parameters after training.
6. Write down the factored conditional probability expression that corresponds to the
graphical Bayesian Network shown below.
7. How do we learn the conditional probability tables(CPT) in Bayesian networks if

information about some variables is missing? How are these variables called?

1. Suppose 10000 patients get tested for flu; out of them, 9000 are actually healthy and 1000
are actually sick. For the sick people, a test was positive for 620 and negative for 380. For
healthy people, the same test was positive for 180 and negative for 8820. Construct a
confusion matrix for the data and compute the accuracy, precision and recall for the data.
2. Given the following data, construct the ROC curve of the data. Compute the AUC.
Thres TP TN FP FN
hold
1 0 25 0 29
2 7 25 0 22
3 18 24 1 11
4 26 20 5 3
5 29 11 14 0
6 29 0 25 0
7 29 0 25 0
3. With an example classification problem, explain the following terms: a) Hyper parameters
b) Training set c) Validation sets d) Bias e) Variance.
4. What is ensemble learning? Can ensemble learning using linear classifiers learn
classification of linearly non-separable sets?
5. Describe boosting. What is the relation between boosting and ensemble learning?
6. Classifier A attains 100% accuracy on the training set and 70% accuracy on the test set.
Classifier B attains 70% accuracy on the training set and 75% accuracy on the test set.
Which one is a better classifier. Justify your answer.
7. What are ROC space and ROC curve in machine learning? In ROC space, which points
correspond to perfect prediction, always positive prediction and always negative
prediction? Why?
8. Suppose there are three classifiers A,B and C. The (FPR, TPR) measures of the three
classifiers are as follows – A (0, 1), B (1, 1) , C (1,0.5). Which can be considered as a
perfect classifier? Justify your answer.
9. What does it mean for a classifier to have a high precision but low recall?
Model Question Paper
QP CODE:
Reg No: _______________
Name: _________________ PAGES : 4
APJ ABDUL KALAM TECHNOLOGICAL UNIVERSITY
FIRST SEMESTER M.TECH DEGREE EXAMINATION, MONTH & YEAR
Course Code: 221TCS009
Course Name: MACHINE LEARNING
Max. Marks: 60 Duration: 2.5 Hours
PART A
Answer All Questions. Each Question Carries 5 Marks
1. Explain the principle of the gradient descent algorithm.

2. In a two-class logistic regression model, the weight vector w = [4, 3, 2, 1, 0]. We
apply it to some object that we would like to classify; the vectorized feature
representation of this object is x = [−2, 0,−3, 0.5, 3]. What is the probability,
according to the model, that this instance belongs to the positive class?
3. Expectation maximization (EM) is designed to find a maximum likelihood setting
of the parameters of model when some of the data is missing. Does the algorithm
converge? If so, do you obtain a locally or globally optimal set of parameters?
4. What is the basic idea of a Support Vector Machine?
5. What is the trade-off between bias and variance? (5x5=25)
Part B
(Answer any five questions. Each question carries 7 marks)
6. Suppose x1, ..., xn are independent and identically distributed(iid) samples (7)
from a distribution with density
Find the maximum likelihood estimate(MLE) for θ.

7. Derive the gradient descent training rule assuming for the target function od = (7)
w0 + w1x1 + ... + wnxn. Define explicitly the squared cost/error function E,
assuming that a set of training examples D is provided, where each training
example d ɛ D is associated with the target output td.
8. Cluster the following eight points representing locations into three clusters: (7)
A1(2, 10), A2(2, 5), A3(8, 4), A4(5, 8), A5(7, 5), A6(6, 4), A7(1, 2), A8(4, 9).
Initial cluster centers are: A1(2, 10), A4(5, 8) and A7(1, 2).
The distance function between two points a = (x1, y1) and b = (x2, y2) is
defined as D(a, b) = |x2 – x1| + |y2 – y1|
Use k-Means Algorithm to find the three cluster centers after the second
iteration.
9. Describe Principal Component Analysis. What criterion does the method (7)
minimize? What is the objective of the method? Give a way to compute the
solution from a matrix X encoding the features.
10. Consider a support vector machine whose input space is 2-D, and the inner (7)
products are computed by means of the kernel K(x, y) = (x.y + 1)2-1 (x.y
denotes the ordinary inner product). Show that the mapping to feature space
that is implicitly defined by this kernel is the mapping to 5-D given by
11. How does random forest classifier work? Why is a random forest better than a (7)
decision tree?
12. Consider a two-class classification problem of predicting whether a (7)
photograph contains a man or a woman. Suppose we have a test dataset of 10
records with expected outcomes and a set of predictions from our
classification algorithm. Compute the confusion matrix, accuracy, precision,
recall, sensitivity and specificity on the following data.
Sl.No. Actual Predicted
1 man woman
2 man man
3 woman woman
4 man man
5 man woman
6 woman woman
7 woman man
8 man man
9 man woman
10 woman woman
Syllabus
Module-1 (Parameter Estimation and Regression) 8 hours
Overview of machine learning: supervised, semi-supervised, unsupervised learning,

reinforcement learning. Basics of parameter estimation: Maximum Likelihood
Estimation(MLE), Maximum a Posteriori Estimation (MAP). Gradient Descent Algorithm,
Batch Gradient Descent, Stochastic Gradient Descent. Regression algorithms: least squares
linear regression, normal equations and closed form solution, Polynomial regression.
Module-2 (Regularization techniques and Classification algorithms) 9 hours
Overfitting, Regularization techniques - LASSO and RIDGE. Classification algorithms:

linear and non-linear algorithms, Perceptrons, Logistic regression, Naive Bayes, Decision
trees. Neural networks : Concept of Artificial neuron, Feed-Forward Neural Network, Back
propagation algorithm.
Module-3 (Unsupervised learning) 8 hours
Unsupervised learning: clustering, k-means, Hierarchical clustering, Principal component analysis,

Density-based spatial clustering of applications with noise (DBSCAN). Gaussian mixture models:
Expectation Maximization (EM) algorithm for Gaussian mixture model.
Module-4 (Support Vector Machine and Graphical Models ) 7 hours
Support vector machines and kernels : Max margin classification, Nonlinear SVM and the
kernel trick, nonlinear decision boundaries, Kernel functions. Basics of graphical models -
Bayesian networks, Hidden Markov model - Inference and estimation.
Module-5 (Evaluation Metrics and Sampling Methods) 8 hours
Classification Performance Evaluation Metrics: Accuracy, Precision, Precision, Recall,

Specificity, False Positive Rate (FPR), F1 Score, Receiver Operator Characteristic (ROC)
Curve, AUC. Regression Performance Evaluation Metrics: Mean Absolute Error (MAE),
Root Mean Squared Error (RMSE), R Squared/Coefficient of Determination. Clustering
Performance Evaluation Metrics: Purity, Jaccard index, Normalized Mutual Information,
Clustering Accuracy, Silhouette Coefficient, Dunn’s Index. Boosting: AdaBoost, gradient
boosting machines. Resampling methods: cross-validation, bootstrap. Ensemble methods:
bagging, boosting, random forests Practical aspects in machine learning: data preprocessing,
overfitting, accuracy estimation, parameter and model selection Bias-Variance tradeoff
Course Plan
No Topics No. of
Lectures
( 40)
1 Module-1 (Parameter Estimation and Regression) 8 hours
1.1 Overview of machine learning: supervised, semi-supervised, 1
unsupervised learning, reinforcement learning.
1.2 Basics of parameter estimation: Maximum Likelihood Estimation(MLE) 1
1.3 Basics of parameter estimation: Maximum Likelihood Estimation(MLE) 1
- Examples
1.4 Basics of parameter estimation: Maximum a Posteriori Estimation (MAP) 1
1.5 Basics of parameter estimation: Maximum a Posteriori Estimation (MAP) 1
- Example
1.6 Gradient Descent Algorithm, Batch Gradient Descent, Stochastic 1
Gradient Descent
1.7 Regression algorithms: least squares linear regression, normal equations 1
and closed form solution
1.8 Polynomial regression 1
2 Module-2 (Regularization techniques and Classification algorithms) 9 hours
2.1 Overfitting, Regularization techniques - LASSO and RIDGE
2.2 Classification algorithms: linear and non-linear algorithms
2.3 Perceptrons
2.4 Logistic regression
2.5 Naive Bayes
2.6 Decision trees
2.7 Neural networks : Concept of Artificial neuron
2.8 Feed-Forward Neural Network
2.9 Back propagation algorithm
3 Module-3 (Unsupervised learning) 8 hours
3.1 Unsupervised learning: clustering, k-means
3.2 Hierarchical clustering
3.3 Principal component analysis
3.4 Density-based spatial clustering of applications with noise (DBSCAN)
3.5 Gaussian mixture models: Expectation Maximization (EM) algorithm for
Gaussian mixture model
3.6 Gaussian mixture models: Expectation Maximization (EM) algorithm for
Gaussian mixture model
4 Module-4 (Support Vector Machine and Graphical Models ) 7 hours
4.1 Support vector machines and kernels : Max margin classification
4.2 Support vector machines: Max margin classification

4.3 Nonlinear SVM and the kernel trick, nonlinear decision boundaries
4.3 Kernel functions
4.5 Basics of graphical models - Bayesian networks
4.6 Hidden Markov model - Inference and estimation
5 Module-5 (Evaluation Metrics and Sampling Methods) 8 hours
5.1 Classification Performance Evaluation Metrics: Accuracy, Precision,
Precision, Recall, Specificity, False Positive Rate (FPR), F1 Score,
Receiver Operator Characteristic (ROC) Curve, AUC
5.2 Regression Performance Evaluation Metrics: Mean Absolute Error
(MAE), Root Mean Squared Error (RMSE), R Squared/Coefficient of
Determination
5.3 Clustering Performance Evaluation Metrics: Purity, Jaccard index,
Normalized Mutual Information, Clustering Accuracy, Silhouette
Coefficient, Dunn’s Index
5.4 Boosting: AdaBoost, gradient boosting machines.
5.5 Resampling methods: cross-validation, bootstrap.
5.6 Ensemble methods: bagging, boosting, random forests
5.7 Practical aspects in machine learning: data preprocessing, overfitting,
accuracy estimation, parameter and model selection
5.8 Bias-Variance tradeoff
Reference Books
1. Christopher Bishop. Neural Networks for Pattern Recognition, Oxford University Press,
1995.
2. Kevin P. Murphy. Machine Learning: A Probabilistic Perspective, MIT Press 2012.

3. Trevor Hastie, Robert Tibshirani, Jerome Friedman, The Elements Of Statistical Learning,
Second edition Springer 2007.
4. Ethem Alpaydin, Introduction to Machine Learning, 2nd edition, MIT Press 2010.
5. Tom Mitchell, Machine Learning, McGraw-Hill, 1997.

221TCS010 MATHEMATICS FOR CATEGORY L T P CREDI
MACHINE LEARNING T
PROGRAM 3 0 0 3
CORE 1
Preamble: Machine learning (ML) is one of the most popular topics of today’s research. It
have applications in almost all the areas of engineering and sciences. Various tools of
machine learning have a rich mathematical theory. Therefore, in order to develop new
algorithms of machine/deep learning, it is necessary to have knowledge of all such
mathematical concepts. The purpose of this course is to introduce mathematical foundations
of basic Machine Learning concepts among learners, on which Machine Learning systems are
built. This course covers Linear Algebra, Vector Calculus, Probability and Distributions,
Optimization those are having strong linkage with machine learning and Machine Learning
problems. Concepts in this course help the learners to understand the mathematical principles
in Machine Learning and aid in the creation of new Machine Learning solutions, understand
& debug existing ones, and learn about the inherent assumptions & limitations of the current
methodologies.
Prerequisite: Nil
Course Outcomes:
After the completion of the course the student will be able to
CO 1 Make use of the concepts, rules and results about matrix algebra, vector spaces,
eigenvalues & eigenvectors and orthogonality & diagonalization to solve
computational problems (Cognitive Knowledge Level: Apply)
CO 2 Perform calculus operations on functions of several variables and matrices, including
partial derivatives and gradients.(Cognitive Knowledge Level: Apply)
CO 3 Utilize the concepts, rules and results about probability, random variables, additive &
multiplicative rules, conditional probability, probability distributions and Bayes’
theorem to find solutions of computational problems. (Cognitive Knowledge Level:
Apply)
CO 4 Train Machine Learning Models using unconstrained and constrained optimization
methods. (Cognitive Knowledge Level: Apply)
CO 5 Illustrate how the mathematical objects - linear algebra, probability, and calculus
can be used to design machine learning algorithms. (Cognitive Knowledge Level:
Apply)
CO 1
CO 2
CO 3
CO 4
CO 5
CO 6
Assessment Pattern
Apply 80%
Analyse 20%
Evaluate
Create
Mark distribution
Total CIE ESE ESE
Marks Duration
100 40 60 2.5 hours

Course based task/test paper questions shall be useful in the testing of knowledge,
skills,vcomprehension, application, analysis, synthesis, evaluation and understanding of
the students.
achievement and maturity of the students in a course, through long answer questions relating
to theoretical/practical knowledge, derivations, problem solving and quantitative evaluation),
with minimum one question from each module of which student should answer any five.
Each question can carry 7 marks.

1. Are the following sets of vectors linearly independent?
2. A set of n linearly independent vectors in Rn forms a basis. Does the set of vectors (2,
4,−3) , (0, 1, 1) , (0, 1,−1) form a basis for R3? Explain your reasons.
3. Consider the transformation T (x, y) = (x + y, x + 2y, 2x + 3y). Obtain ker T and use this to
calculate the nullity. Also find the transformation matrix for T.
4.
Find the characteristic equation, eigenvalues, and eigenspaces corresponding to each eigenvalue of
the following matrix
5. Diagonalize the following matrix, if possible
6. Find the singular value decomposition (SVD) of the following matrix
1. For a scalar function f(x, y, z ) = x2 +3y2 +2z2, find the gradient and its magnitude at the
point (1, 2, -1).
2. Find the maximum and minimum values of the function f(x, y) = 4x + 4y - x2 - y2 subject
to the condition x2 + y2 <= 2.
3. Suppose you were trying to minimize f(x, y) = x2+ 2y + 2y2. Along what vector should
you travel from (5, 12)?
4. Find the second order Taylor series expansion for f(x, y) = (x + y)2 about (0 , 0).
1. Let J and T be independent events, where P(J)=0.4 and P(T)=0.7.
i. Find P(J∩T)
ii. Find P(J∪T)

iii. Find P(J∩T′)
2. Let A and B be events such that P(A)=0.45 , P(B)=0.35 and P(A∪B)=0.5. Find P(A∣B).
3. A random variable R has the probability distribution as shown in the following table:
i.
i. Given that E(R)=2.85, find a and b.
ii. Find P(R>2).
4. A biased coin (with probability of obtaining a head equal to p > 0) is tossed repeatedly and
independently until the first head is observed. Compute the probability that the first head
appears at an even numbered toss.
5. Two players A and B are competing at a trivia quiz game involving a series of questions.
On any individual question, the probabilities that A and B give the correct answer are p and
q respectively, for all questions, with outcomes for different questions being independent.
The game finishes when a player wins by answering a question correctly. Compute the
probability that A wins if
i. A answers the first question,
ii. B answers the first question.
6. A coin for which P(heads) = p is tossed until two successive tails are obtained. Find the
probability that the experiment is completed on the nth toss.
7. You roll a fair dice twice. Let the random variable X be the product of the outcomes of the
two rolls. What is the probability mass function of X? What are the expected value and the
standard deviation of X?
8. While watching a game of Cricket, you observe someone who is clearly supporting
Mumbai Indians. What is the probability that they were actually born within 25KM of
Mumbai? Assume that:
 the probability that a randomly selected person is born within 25KM of Mumbai is
1/20;
 the chance that a person born within 25KMs of Mumbai actually supports MI is
7/10 ;
 the probability that a person not born within 25KM of Mumbai supports MI with
probability 1/10.
9. Let Z1 and Z2 be independent random variables each having the standard normal
distribution. Define the random variables X and Y by X = Z1 + 3Z2 and Y = Z1 + Z2. Argue
that the joint distribution of (X, Y) is a bivariate normal distribution. What are the
parameters of this distribution?
10. Given a continuous random variable x, with cumulative distribution function Fx(x), show
that the random variable y = Fx(x) is uniformly distributed.
1. Find the extrema of f(x, y) = x subject to g(x, y) = x2 + 2y2 = 3.
2. Maximize the function f(x, y, z) = xy + yz + xz on the unit sphere g(x, y, z) = x2 + y2

+ z2 = 1.
3. Provide necessary and suffcient conditions under which a quadratic optimization

problem be written as a linear least squares problem.
4. Consider the univariate function f(x) = x3 + 6x2 - 3x - 5. Find its stationary points
and indicate whether they are maximum, minimum, or saddle points.
5. Consider the update equation for stochastic gradient descent. Write down the update
when we use a mini-batch size of one.
6. Consider the function
i. Is f(x) a convex function? Justify your answer.

ii. Is (1 , -1) a local/global minimum? Justify your answer.
7. Is the function f(x, y) = 2x2 + y2 + 6xy - x + 3y - 7 convex, concave, or neither?

Justify your answer.
8. Consider the following convex optimization problem
Subject to the constraint x + y >= 4, x, y >= 1.
Derive an explicit form of the Lagrangian dual problem.
9. Solve the following LP problem with the simplex method.
subject to the constraints
1. What is a kernel? What is a dot product? Give examples of kernels that are valid dot
products.
2. What is Principal Component Analysis (PCA)? Which eigen value indicates the direction of
largest variance? In what sense is the representation obtained from a projection onto the
eigen directions corresponding the the largest eigen values optimal for data reconstruction?
3. Suppose that you have a linear support vector machine (SVM) binary classifier. Consider a
point that is currently classified correctly, and is far away from the decision boundary. If you
remove the point from the training set, and re-train the classifier, will the decision boundary
change or stay the same? Explain your answer in one sentence.
4. Suppose you have n independent and identically distributed (i.i.d) sample data points x1, ... ,
xn. These data points come from a distribution where the probability of a given datapoint x is
Prove that the MLE estimate of parameter is the sample mean.

QP CODE:
Reg No: _______________
Name: _________________ PAGES : 4
Course Name: MATHEMATICS FOR MACHINE LEARNING
Max. Marks : 60 Duration: 2.5 Hours
PART A
1. Show that with the usual operation of scalar multiplication but with addition on
reals given by x # y = 2(x + y) is not a vector space.
r 2 2
2. Let f(x, y, z) = xye , where r = x +z -5. Calculate the gradient of f at the point
(1, 3, -2).
3. Let X be a continuous random variable with probability density function on

0 <= x <= 1 defined by f(x) = 3x2. Find the pdf of Y = X2.
4. Explain the principle of the gradient descent algorithm.
5. Explain the concept of a Kernel function in Support Vector Machines. Why are
kernels so useful? What properties a kernel should posses to be used in an SVM? (5x5=25)
Part B
6. Prove that all vectors orthogonal to [2, −3, 1]T forms a subspace W of (7)
3
R . What is dim (W) and why?
7. Find the rank-1 approximation of (7)

8. A skier is on a mountain with equation z = 100 – 0.4x2 – 0.3y2, where z (7)

denotes height.
i. The skier is located at the point with xy-coordinates (1 , 1), and wants
to ski downhill along the steepest possible path. In which direction
(indicated by a vector (a , b) in the xy-plane) should the skier begin
skiing.
ii. The skier begins skiing in the direction given by the xy-vector (a , b)
you found in part (i), so the skier heads in a direction in space given
by the vector (a , b , c). Find the value of c.
9. Find the linear approximation to the function f(x,y) = 2 - sin(-x - 3y) at the (7)
point (0 , π), and then use your answer to estimate f(0.001 , π).
10 There are two bags. The first bag contains four mangos and two apples; the (7)
second bag contains four mangos and four apples. We also have a biased coin,
which shows “heads” with probability 0.6 and “tails” with probability 0.4. If
the coin shows “heads”. we pick a fruit at
random from bag 1; otherwise we pick a fruit at random from bag 2. Your
friend flips the coin (you cannot see the result), picks a fruit at random from
the corresponding bag, and presents you a mango.
What is the probability that the mango was picked from bag 2?
11 Derive the gradient descent training rule assuming that the target function is (7)
represented as od = w0 + w1x1 + ... + wnxn. Define explicitly the cost/error
function E, assuming that a set of training examples D is provided, where
each training example d ∈ D is associated with the target output td.
12 Suppose that we are fitting a Gaussian mixture model for data items (7)
consisting of a single real value, x, using K = 2 components. We have N = 5
training cases, in which the values of x are as 5, 15, 25, 30, 40. Using the EM
algorithm to find the maximum likeihood estimates for the model parameters,
what are the mixing proportions for the two components, π1 and π2, and the
means for the two components, μ1 and μ2. The standard deviations for the two
components are fixed at 10.
Suppose that at some point in the EM algorithm, the E step found that the
responsibilities of the two
components for the five data
items were as follows:
What values for the parameters π1, π2 , μ1, and μ2 will be found in the next M
step of the algorithm?
Syllabus
Module – 1 (LINEAR ALGEBRA)
Vectors in Machine Learning, Basics of Matrix Algebra, Vector Spaces, Subspaces, Linear
Independence, Basis and Rank, Linear Transformations, Norms and Innerproducts,
Orthogonal Complement and Projection Mapping, Eigenvalues and Eigenvectors, Eigen
decomposition and Diagonalization, Singular Value Decomposition, Low-rank Matrix
Approximation.
Module – 2 (VECTOR CALCULUS)

Basics concepts of Calculus, Differentiation of Univariate Functions Partial Differentiation
and Gradients, Gradients of Vector Valued Functions and Matrices, Jacobian, Chain rule,
Change of variables, Linearization and Multivariate Taylor Series.
Module – 3 (Probability and Probability Distributions)

Review of Probability, Discrete and Continuous Probability Spaces, Sum Rule, Product Rule,
Independence, Bayes’ Theorem. Random Variables, Expectation and variance, Discrete and
continuous distribution functions, Joint probability and Covariance.
Module – 4 (Optimization)
Optimization Using Gradient Descent, Constrained Optimization and Lagrange Multiplier
Method, Concepts of duality, Convex Optimization, Linear Programming, Quadratic
Programming.
Module – 5 (Central Machine Learning Problems)
Linear Regression - Bayesian Linear Regression - Maximum Likelihood as Orthogonal

Projection.
Dimensionality Reduction with Principal Component Analysis - Maximum Variance
Perspective, Projection Perspective.
Classification with Support Vector Machines - Separating Hyperplanes, Primal Support
Vector Machine, Dual Support Vector Machine, Kernels.
Corse Plan (For 3 credit courses, the content can be for 40 hrs and for 2 credit courses, the
content can be for 26 hrs. The audit course in third semester can have content for 30 hours).
No Topic No. of Lectures

1 LINEAR ALGEBRA(12 Hours)
1.1 Vectors in Machine Learning, Basics of Matrix Algebra 1
1.2 Vector Spaces 1
1.3 Subspaces 1
1.4 Linear Independence 1
1.5 Basis and Rank 1
1.6 Linear Transformations 1
1.7 Norms and Innerproducts 1
1.8 Orthogonal Complement and Projection Mapping 1
1.9 Eigenvalues and Eigenvectors 1
1.10 Eigen decomposition and Diagonalization 1
1.11 Singular Value Decomposition 1
1.12 Low-rank Matrix Approximation 1
2 VECTOR CALCULUS(5 Hours)
2.1 Basics concepts of Calculus, Differentiation of Univariate 1
Functions
2.2 Partial Differentiation and Gradients, Gradients of Vector Valued 1
Functions and Matrices
2.3 Jacobian, Chain rule 1
2.4 Change of variables 1
2.5 Linearization and Multivariate Taylor Series 1
3 Probability and Probability Distributions(9 Hours)

3.1 Review of Probability, Discrete and Continuous Probability 1
Spaces
3.2 Sum Rule, Product Rule 1
3.3 Independence 1
3.4 Bayes’ Theorem 1
3.5 Random Variables 1
3.6 Expectation and variance 1
3.7 Discrete and continuous distribution functions 1
3.8 Joint probability 1
3.9 Covariance 1
4 Optimization(6 Hours)
4.1 Optimization Using Gradient Descent 1
4.2 Constrained Optimization and Lagrange Multiplier Method, 1
Concepts of duality
4.3 Constrained Optimization and Lagrange Multiplier Method, 1
Concepts of duality
4.4 Convex Optimization 1
4.5 Linear Programming 1
4.6 Quadratic Programming. 1
5 CENTRAL MACHINE LEARNING PROBLEMS(8 Hours)
5.1 Linear Regression - Bayesian Linear Regression 1
5.2 Linear Regression - Maximum Likelihood as Orthogonal 1
Projection.
5.3 Dimensionality Reduction with Principal Component Analysis - 1
Maximum Variance Perspective
5.4 Dimensionality Reduction with Principal Component Analysis - 1
Projection Perspective.
5.5 Support Vector Machines - Separating Hyperplanes, 1
5.6 Primal Support Vector Machine 1
5.7 Dual Support Vector Machine 1
5.8 Kernels. 1
Reference Books
1. Mathematics for Machine Learning by Marc Peter Deisenroth, A. Aldo Faisal, and
Cheng Soon Ong published by Cambridge University Press (freely available at
https://mml – book.github.io)
2. Linear Algebra and Its Applications, 4th Edition by Gilbert Strang
3. Linear Algebra Done Right by Axler, Sheldon, 2015 published by Springer
4. Introduction to Applied Linear Algebra by Stephen Boyd and Lieven Vandenberghe,

2018 published by Cambridge University Press
5. Convex Optimization by Stephen Boyd and Lieven Vandenberghe, 2004 published by

Cambridge University Press
6. Pattern Recognition and Machine Learning by Christopher M Bishop, 2006, published

by Springer
7. Learning with Kernels – Support Vector Machines, Regularization, Optimization, and

Beyond by Bernhard Scholkopf and Smola, Alexander J Smola, 2002, bublished by
MIT Press
8. Information Theory, Inference, and Learning Algorithms by David J. C MacKay,

2003 published by Cambridge University Press
9. Machine Learning: A Probabilistic Perspective by Kevin P Murphy, 2012 bublished

by MIT Press.
10. The Nature of Statistical Learning Theory by Vladimir N Vapnik, 2000, published by
Springer
CODE COURSE NAME CATEGORY L T P CREDIT
221TCS011 FOUNDATIONS OF AI PROGRAM 3 0 0 3
CORE 2
Preamble:
The course introduces the fundamental principles of intelligent systems. This course covers
the characteristics of intelligent systems, knowledge representation schemes, logic and
inference mechanisms. The course helps the learner to develop the design of self-learning
systems along with some of their typical applications in the emerging scenario where the
business world is being transformed by the progress made in machine learning.
Course Outcomes:
After the completion of the course the student will be able to:
CO 1 Analyse the concepts of intelligent systems and their architecture.(Cognitive

knowledge level: Analyse)
CO 2 Apply the most appropriate algorithm for search and reasoning within an AI problem
domain. (Cognitive knowledge level: Apply)
CO 3 Represent AI domain knowledge using logic systems and use inference techniques
for reasoning in intelligent systems. (Cognitive knowledge level: Apply)
CO 4 Solve Constraint Satisfaction Problems using search techniques. (Cognitive
knowledge level: Apply)
CO 5 Apply suitable learning techniques to develop intelligent systems (Cognitive
knowledge level: Apply)
Program Outcomes ( PO )
course.
PO1: An ability to independently carry out research/investigation and development work in

engineering activities by interacting with the engineering fraternity and with society at
large.
PO3: An ability to demonstrate a degree of mastery over the area as per the specialization of
the program. The mastery should be at a level higher than the requirements in the
appropriate bachelor program
PO4: An ability to apply stream knowledge to design or develop solutions for real world
problems by following the standards
PO5: An ability to identify, select and apply appropriate techniques, resources and state-of-
the-art tool to model, analyse and solve practical engineering problems.
PO6: An ability to engage in life-long learning for the design and development related to the
stream related problems taking into consideration sustainability, societal, ethical and
environmental aspects
PO7: An ability to develop cognitive load management skills related to project management
and finance which focus on Entrepreneurship and Industry relevance.
CO 1
CO 2
CO 3
CO 4
CO 5
Assessment Pattern
Apply 50-80%
Analyse 20-40%
Evaluate Assess using
Assignments/Projects
Create Assess using
Assignments/Projects
Mark distribution
Total CIE ESE ESE

Marks Duration
100 40 60 2.5 hours


students.

1. Explain about the basic types of agent programs in intelligent systems.
2. For the following activities, give a PEAS description of the task environment and
characterize it in terms of the task environment properties.
a) Playing soccer.
b) Bidding on an item at an auction.

1. Differentiate between uninformed and informed search strategies in intelligent systems.
2. Illustrate the working of Minimax search procedure.

1. Prove, or find a counter example to, the following assertion:
If α |= γ or β |= γ (or both) then (α ∧ β) |= γ
2. For each pair of atomic sentences, find the most general unifier if it exists:
a) P (A, B, B), P (x, y, z).
b) Q(y, G(A, B)), Q(G(x, x), y).

1. Solve the following crypt arithmetic problem by hand, using the strategy of backtracking
with forward checking and the MRV & least-constraining-value heuristics.

1. Consider the following data set comprised of three binary input attributes (A1 , A2 , and
A3 ) and one binary output
Use the DECISION-TREE-LEARNING algorithm to learn a decision tree for these data.
Show the computations made to determine the attribute to split at each node.
2. What is multivariate linear regression? Explain.

QP CODE:
Reg No: _______________
Name: _________________ PAGES : 4
Course Name: FOUNDATIONS OF AI
PART A
1. What is a rational agent? Explain with the help of a suitable example.
2. Differentiate between informed search and uninformed search.

3. a. What are the components of a Constraint Satisfaction Problem? Illustrate
with an example.
b. Formulate the following problem as a CSP. Class scheduling: There is a

fixed number of professors and classrooms, a list of classes to be offered,
and a list of possible time slots for classes. Each professor has a set of
classes that he or she can teach.
4. Represent the following assertion in propositional logic:

“A person who is radical (R) is electable (E) if he/she is conservative (C), but
otherwise is not electable.”
5. Describe the various forms of learning? (5x5=25)
Part B
6. (a) Explain the structure Goal-based agents and Utility-based agents with the (7)
help of diagrams.
(b) For the following activities, give a PEAS description of the task environment
and characterize it in terms of the task environment properties.
a) Playing soccer
b) Bidding on an item at an auction.
7. (a) Discuss about any five applications of AI. (7)
(b) Explain Best First Search algorithm. How does it implement heuristic
search?
8. (a) Describe any four uninformed search strategies. (7)
(b) Write and explain A* search algorithm.
9. (a) Explain the components of a well-defined AI problem? Write the standard (7)
formulation of 8-puzzle problem.
(b) What is local consistency in CSP constraint propagation? Explain different
types local consistencies.
10. (a) Convert the following sentences into first order logic: (7)
Everyone who loves all animals is loved by someone.
Anyone who kills an animal is loved by no one.
Jack loves all animals.
Either Jack or Curiosity killed the cat, who is named Tuna.
Did Curiosity kill the cat?
(b) Give a resolution proof to answer the question “Did Curiosity kill the cat? “
11. (a) Prove or find a counter example to the following assertion in propositional (7)
logic:
If α |= (β ∧ γ) then α |= β and α |= γ.
(b) How is best hypothesis selected from alternatives?
12. (a) Explain Univariate Linear Regression. (7)
(b) Consider the following data set comprised of two binary input attributes (A1
and A2) and one binary output.
Use the DECISION-TREE-LEARNING algorithm to learn a decision tree for

these data. Show the computations made to determine the attribute to split at
each node.
Syllabus
Module – 1 (Introduction)
Introduction – What is Artificial Intelligence(AI)? The Foundations of AI, History of AI,

Applications of AI. Intelligent Agents – Agents and Environments, Good behavior: The
concept of rationality, nature of Environments, Structure of Agents.
Module – 2 (Problem Solving)
Solving Problems by Searching-Problem solving Agents, Example problems, Searching for

solutions, Uninformed search strategies, Informed search strategies, Heuristic functions.
Module - 3 (Search in Complex environments)
Adversarial search - Games, Optimal decisions in games, The Minimax algorithm, Alpha-
Beta pruning. Constraint Satisfaction Problems – Defining CSP, Constraint Propagation-
inference in CSPs, Backtracking search for CSPs, Structure of CSP problems.
Module - 4 (Knowledge Representation and Reasoning)
Logical Agents – Knowledge based agents, Logic, Propositional Logic, Propositional

Theorem proving, Agents based on Propositional Logic. First Order Predicate Logic – Syntax
and Semantics of First Order Logic, Using First Order Logic, Knowledge representation in
First Order Logic. Inference in First Order Logic – Propositional Vs First Order inference,
Unification and Lifting, forward chaining, Backward chaining, Resolution.
Module - 5 (Machine Learning)

Learning from Examples – Forms of Learning, Supervised Learning, Learning Decision
Trees, Evaluating and choosing the best hypothesis, Regression and classification with Linear
models.
Course Plan

1 Module 1(Introduction to Artificial Intelligence (AI))
1.1 What is AI 1
1.2 History of AI 1
1.3 Applications of AI 1
1.4 Intelligent Agents – Agents and Environments 1
1.5 Good behaviour: The concept of rationality 1
1.6 Nature of Environments 1
1.7 Structure of Agents 1
2 Module 2 (Problem Solving)
2.1 Solving Problems by searching 1
2.2 Problem solving Agents 1
2.3 Example problems 1
2.4 Searching for solutions 1
2.5 Uninformed search strategies 1
2.6 Informed search strategies 1
2.7 Heuristic functions 1
3 Module 3 (Search in Complex environments)
3.1 Adversarial search – Games, Optimal decisions in games 2
3.2 The Minimax algorithm 1
3.3 Alpha-Beta pruning 1
3.4 Constraint Satisfaction Problems – Defining CSP 1
3.5 Constraint Propagation- inference in CSPs 1
3.6 Local search for CSPs 1
3.7 Structure of CSP problems 1
4 Module 4 (Knowledge Representation and Reasoning)
4.1 Logical Agents – Knowledge based agents 1
4.2 Propositional Logic 1
4.3 Propositional Theorem proving 1
4.4 Agents based on Propositional Logic. 1
4.5 First Order Predicate Logic – Syntax and Semantics of First Order 1
Logic
4.6 Knowledge engineering in in First Order Logic 1
4.7 Inference in First Order Logic – Propositional Vs First Order 1
inference
4.8 Unification and Lifting 1
4.9 Forward chaining, Backward chaining 1
4.10 Resolution 2
5 Module 1 (Machine Learning)
5.1 Learning from Examples- Forms of Learning 2
5.2 Supervised Learning 1
5.3 Learning Decision Trees 1
5.4 Evaluating and choosing the best hypothesis 1
5.5 Regression and classification with Linear models. 2
Reference Books
1. Stuart Russell and Peter Norvig. Artificial Intelligence: A Modern Approach, 3rd Edition.
Prentice Hall.
2. Nilsson N.J., Artificial Intelligence - A New Synthesis, Harcourt Asia Pvt. Ltd.
3. Luger, George F. Artificial intelligence: structures and strategies for complex problem
solving. Pearson education.
4. Patterson, Dan. Introduction to artificial intelligence and expert systems. Prentice-Hall.

221ECS044 DATA STRUCTURES AND PROGRAM 3 0 0 3
ALGORITHMS ELECTIVE 1
Preamble:
The course introduces to students the basic and advanced data structures, as well as various types of
algorithms. The goal of this course is to provide a solid background in the design and analysis of the
major classes of algorithms. This course will enable the learners to develop their own algorithm for a
given computational task and to compare and contrast their performance.
Course Outcomes:
CO 1 Analyze a given algorithm and express its time complexity in asymptotic notations.
(Cognitive Level: Analyse)
CO 2 Develop applications using data structures like array, stack, queue and linked list. (Cognitive
Level: Apply)
CO 3 Illustrate Binary tree and Graph related algorithms and their applications. (Cognitive Level:
Apply)
CO 4 Understand the relevance of amortized analysis and application. (Cognitive Level:
Understand)
CO 5 Illustrate advanced data structures like Binomial heap, Fibonacci heap, Disjoint set and
string matching algorithms. (Cognitive Level: Apply)
CO 6 Illustrate network flow algorithms and applications. (Cognitive Level: Apply)
CO 7 Design, develop and implement software using advanced data structures and algorithms.
(Cognitive Level: Create)
Program Outcomes ( PO)
course.

engineering activities by interacting with the engineering fraternity and with society at
large.
the program. The mastery should be at a level higher than the requirements in the
appropriate bachelor program
PO5: An ability to identify, select and apply appropriate techniques, resources and state-of-the-
art tool to model, analyse and solve practical engineering problems.
CO 1 ✔️ ✔️ ✔️ ✔️ ✔️
CO 2 ✔️ ✔️ ✔️ ✔️ ✔️
CO 3 ✔️ ✔️ ✔️ ✔️
CO 4 ✔️ ✔️ ✔️ ✔️
CO 5 ✔️ ✔️ ✔️ ✔️
CO 6 ✔️ ✔️ ✔️ ✔️ ✔️ ✔️ ✔️
CO 7 ✔️ ✔️ ✔️ ✔️ ✔️ ✔️ ✔️
Assessment Pattern
Apply 50-80%
Analyse 20-40%
Evaluate Evaluate by
Assignments/Project
Create Evaluate by
Assignments/Project
Mark distribution
Total CIE ESE ESE Duration

Marks
100 40 60 2.5 hours

Preparing a review article based on peer reviewed original publications (minimum 10

publications shall be referred) : 15 marks
Course based task/Seminar/Data collection and interpretation : 15 marks
Test paper (1 number) : 10 marks
students.

1. What is the need of asymptotic analysis in calculating time complexity? What are the notations
used for asymptotic analysis?
2. Write the insertion sort algorithm and analyse the time complexity of the algorithm.
3. Calculate the time complexity for addition of two matrices.
1. Explain stack operations and its applications.
2. Compare singly linked list and doubly linked list.
3. Write the algorithm for deleting an element from array data structure.
1. Construct the minimum spanning tree for the given graph using Kruskal’s algorithm.
2. Create a Binary search Tree with nodes representing the following sequence 14, 15, 4, 18, 9, 16, 20,
17, 3, 7, 5, 2 and perform inorder, preorder and postorder traversals on the above tree and print the
output.
3. Write down the BFS algorithm and analyse its time complexity. Perform BFS traversal on the given
graph starting from node 0.
1. Explain how the accounting method of amortized analysis can be applied to stack operations.
2. Suppose we perform a sequence of n operations on a data structure in which the ith operation costs i
if i is an exact power of 2, and 1 otherwise. Use aggregate analysis to determine the amortized cost
per operation.
3. What is the total cost of executing n of the stack operations PUSH , POP , and MULTIPOP ,
assuming that the stack begins with s0 objects and finishes with sn objects? Use potential method.
1. Analyze the time complexity of decrease-key operation of Fibonacci heap.
2. Illustrate extract-min operation of Binomial heap.
3. Illustrate Knuth-Morris-Pratt algorithm.
1. Show the execution of the Edmonds-Karp algorithm on the given flow network (source: A and
sink: J).
2. In the following figure, compute flow across the cut ({s, v2, v4}, {v1, v3, t}). What is the capacity of
this cut?
QP CODE:
Reg No: _______________
Name: _________________ PAGES : 4
Course Code: 221ECS044
Course Name: DATA STRUCTURES AND ALGORITHMS
PART A
1. Write an algorithm to reverse the nodes of a doubly linked list.
2. Apply Dijkstra’s algorithm over the following graph to obtain single

source shortest path (source vertex is A).
3. Explain accounting method of amortized analysis with a suitable
example.
4. Explain the algorithm for uniting two binomial heaps and analyse
the running time.
5. Maximum matching in a bipartite graph G corresponds to a

maximum flow in its corresponding flow network G’. Comment on (5x5=25)
this statement. Explain how maximum flow problem can be used to
solve maximum bipartite matching problem.
Part B
6. (a) Write an algorithm to insert a node into a singly linked list. (4)
(b) Write the algorithm for queue data structure operations. (3)
7. (a) Apply topological sort over the following directed graph. (4)
(b) Write an algorithm to insert a node into a binary search tree. (3)
8. Write Prim’s algorithm. Construct the minimum spanning tree for the (7)
given graph using Prim’s algorithm.
9. Write and illustrate Knuth-Morris-Pratt algorithm using given (7)
text T = AABAACAADAABAABA and pattern P = AABA.
10 Explain extract minimum operation on Fibonacci heap. Apply (7)

. the algorithm over the following heap and show the result.
11. (a) Explain how disjoint set data structure is used to find connected (4)
components on an undirected graph.
(b) Explain the properties of Fibonacci heap. (3)
12. Write Ford-Fulkerson algorithm and apply on the following (7)

network. Also obtain minimum cut across the network.
Syllabus
Module – 1 (Linear data structures)
Criteria for Analysing Algorithms, Time and Space Complexity - Best, Worst and Average
Case Complexities, Asymptotic Notations - Big-Oh (O), Big- Omega (Ω), Big-Theta (Θ),
Complexity Analysis of Simple Algorithms.
Basic data structures: Array, Stack, Queue, Linked list, Doubly Linked List.
Module – 2 (Non-linear data structures)

Binary tree – properties, Tree traversals, Binary search tree – operations. Graphs,
Representation of Graphs, Depth First Search, Breadth First Search, Topological sort.
Minimum Cost Spanning Tree Computation- Kruskal’s Algorithm, Prim’s Algorithm. Single
Source Shortest Path Algorithm - Dijkstra’s Algorithm.
Module – 3 (Amortized analysis and String matching)
Amortized analysis – aggregate analysis, accounting method, potential method.
String matching – introduction, Rabin-Karp algorithm, Knuth-Morris-Pratt algorithm.
Module – 4 (Advanced data structures)

Introduction to binary heap operations, Binomial tree and heap, Binomial heap operations,
Fibonacci heap structure, Fibonacci heap operations, Disjoint set – overview, linked list
representation, disjoint set forests.
Module – 5 (Network flow)

Network flow properties, examples, residual network, augmenting path, cut of network,
maxflow-mincut theorem, Ford-Fulkerson algorithm, Edmonds-Karp algorithm, maximum
bipartite matching.
Course Plan (For 3 credit courses, the content can be for 40 hrs and for 2 credit courses, the
1 Module–1 (Linear data structures)
1.1 Criteria for Analysing Algorithms, Time and Space Complexity - 1

Best, Worst and Average Case Complexities
1.2 Asymptotic Notations - Big-Oh (O), Big- Omega (Ω), Big-Theta 1

(Θ)
1.3 Complexity Analysis of Simple Algorithms 1
1.4 Basic data structures: Array, Stack 1
1.5 Queue 1
1.6 Linked list 1
1.7 Doubly Linked List 1
2 Module–2 (Non-linear data structures)
2.1 Binary tree – properties, Tree traversals 1
2.2 Binary search tree – operations 1
2.3 Graphs, Representation of Graphs 1

2.4 Depth First Search, Breadth First Search, Topological sort 1
2.5 Minimum Cost Spanning Tree Computation- Kruskal’s Algorithm 1
2.6 Prim’s Algorithm 1
2.7 Single Source Shortest Path Algorithm - Dijkstra’s Algorithm 1
3 Module–3 (Amortized analysis and String matching)
3.1 Amortized analysis – aggregate analysis 1
3.2 Accounting method 1
3.3 Potential method 1
3.4 String matching – introduction 1
3.5 Rabin-Karp algorithm 1
3.6 Knuth-Morris-Pratt algorithm (1) 1
3.7 Knuth-Morris-Pratt algorithm (2) 1
4 Module–4 (Advanced data structures)
4.1 Introduction to binary heap operations 1
4.2 Binomial tree and heap 1
4.3 Binomial heap operations (1) 1
4.4 Binomial heap operations (2) 1
4.5 Fibonacci heap structure 1
4.6 Fibonacci heap operations (1) 1
4.7 Fibonacci heap operations (2) 1
4.8 Disjoint set – operations, linked list representation 1
4.9 Disjoint set forests 1
5 Module–5 (Network flow)
5.1 Network flow properties, examples 1

5.2 Residual network, augmenting path, cut of network 1
5.3 Maxflow-mincut theorem 1
5.4 Ford-Fulkerson algorithm 1
5.5 Edmonds-Karp algorithm 1
5.6 Maximum bipartite matching 1
Reference Books
1. T. H. Cormen, C. E. Leiserson, R. L. Rivest and C. Stein, “Introduction to algorithms”,

Prentice-hall of India Private Limited, New Delhi, 2022.
2. Gilles Brassard and Paul Bratley, “Fundamentals of algorithms”, Prentice-hall of India

Private Limited, New Delhi,2015.
3. Ellis Horowitz, Sartaj Sahni, Sanguthevar Rajasekaran, “Fundamentals of Computer

Algorithms”, 2nd Edition, Orient Longman Universities Press (2008).
4. Sara Baase and Allen Van Gelder, “Computer Algorithms, Introduction to Design and
Analysis”, 3rd Edition, Pearson Education (2009).
CODE CYBER PHYSICAL SYSTEMS CATEGORY L T P CREDIT
221ECS045 PROGRAM 3 0 0 3
ELECTIVE 1
Preamble:
The course introduces the physical systems that are tightly coupled by software to be used in
safety critical domains including automotive, healthcare, energy and power and industrial
automation. This course helps the learners to design and implement cyber physical systems
according to application needs. This course provides an overview of the underlying design
principles for the future cyber physical systems and discusses the research problems in
different application domains.
Course Outcomes: The COs shown are only indicative. For each course, there can be 4 to 6
COs.
CO 1 Analyse the need and purpose of the various components of Cyber Physical Systems.
(Cognitive Knowledge Level : Analyse)
CO 2 Interpret the cross disciplinary interaction of cyber and Physical systems. (Cognitive
Knowledge Level : Apply)
CO 3 Develop concepts, logics towards solving real-world problems in research and

industry. (Cognitive Knowledge Level : Apply)
CO 4 Design real time CPS systems for real world research problems (Cognitive Knowledge
Level : Create)

course.

bachelor program
aspects
CO 1
CO 2
CO 3
CO 4
Assessment Pattern
Bloom’s Category End Semester Examination (%)
Apply 60-80%
Analyse 20-40%
Evaluate Evaluate by Assignments/Project
Create Evaluate by Assignments/Project
Mark distribution
Marks
100 40 60 2.5 hours

Continuous Internal Evaluation: 40 marks
i. Preparing a review article based on peer reviewed original publications (minimum 10

ii. Course based task / Seminar/ Data collection and interpretation : 15 marks
iii. Test paper (1 number) : 10 marks
comprehension, application, analysis, synthesis, evaluation and understanding of the students.

The end semester examination will be conducted by the respective College.
There will be two parts; Part A and Part B.
Part A will contain 5 numerical/short answer questions with 1 question from each module, having 5
marks for each question. Students should answer all questions. Part B will contain 7 questions (such
questions shall be useful in the testing of overall achievement and maturity of the students in a
course, through long answer questions relating to theoretical/practical knowledge, derivations,
problem solving and quantitative evaluation), with minimum one question from each module
of which student should answer any five. Each question can carry 7 marks
Note: The marks obtained for the ESE for an elective course shall not exceed 20% over the average
ESE mark % for the core courses. ESE marks awarded to a student for each elective course shall be
normalized accordingly.
For example if the average end semester mark % for a core course is 40, then the maximum eligible
mark % for an elective course is 40+20 = 60 %.

1. Consider a product that is expected to appear on the market in the near future. For this
product, describe the issues that should be addressed at the idea, model, prototype,
and product stages.
2. Identify need for CPS in a smart city environment. Find out the requirements

1. Demonstrate the design principles of CPS
2. Illustrate the implementation of the interactions between cyber and physical systems
3. Differentiate CPS and IoT

1. Demonstrate the hardware platforms used by different CPS.
2. Illustrate the reactive components in synchronous model
3. Illustrate the different wireless technologies used in CPS

1. Design a CPS model for medical CPS
2. Design attack model for any one CPS system.

QP CODE:
Reg No: _______________
Name: _________________ PAGES : 4
Course Name: CYBER PHYSICAL SYSTEMS
PART A
1. List out the basic features of CPS.

2. Compare sensors and actuators.
3. How the leader is elected in a CPS model.
4. Define the attack model used in CPS.
5. List out the CPS challenges in smart city environment. (5x5=25)
Part B
6. (a) Write down the basic principles in the design of CPS (7)
(b) List out the challenges in the design of CPS
7. (a) Describe the various hardware platforms for CPS (7)
(b) Explain the different wireless technologies used in CPS
8. (a) Explain about the reactive components in synchronous model (7)
9. (a) List out the security and privacy issues in CPS (7)
Explain the cyber security requirements in CPS
10. (a) What are the design primitives in asynchronous process (7)
11. (a) Explain in detail about CPS in smart energy (7)
12. (a) Define attack model used in medical CPS (7)
Syllabus
Module 1 (Introduction)
Cyber Physical Systems, Key Features of CPS, CPS in real world, Basic principles in the
design of CPS, CPS Design Considerations, Requirements, Challenges, Case Study
Module 2 (Cyber Physical System Platforms)

Cyber physical systems platforms, Hardware platforms for CPS, Processors, Sensors and
Actuators, Wireless technologies for CPS
Module 3 (Cyber Physical System Models and Behaviours)

Synchronous and Asynchronous Models, Reactive Components, Components Properties,
Synchronous Designs and Circuits, Asynchronous Processes and operations, Design
Primitives in Asynchronous Process, Coordination Protocols in Asynchronous Process,
Leader Election, Reliable Transmission
Module 4 (Security and Privacy in Cyber Physical Systems)

Introduction to CPS securities, Security and privacy issues in CPS, Basic techniques in CPS
security, Cyber security requirements, Attack model and Counter measures, Case study
Module 5 (CPS Application)

Tools and programming frameworks for CPS, Medical CPS, Energy CPS, CPS in Smart
cities
Course Plan
No Topic No. of Lectures (

40 hrs)
1 Module 1 (Introduction) 8 Hours
1.1 Cyber Physical Systems 1 hour
1.2 Key Features of CPS 1 hour
1.3 CPS in real world 1 hour
1.4 Basic principles in the design of CPS 1 hour
1.5 CPS Design Considerations 1 hour
1.6 Requirements, Challenges 1 hour
1.7 Case Study 2 hours

2 Module 2 (Cyber Physical System Platforms) 8 Hours
2.1 Cyber physical systems platforms 2 hours
2.2 Hardware platforms for CPS 2 hours
2.3 Processors, Sensors and Actuators 2 hours
2.4 Wireless technologies for CPS 2 hours
3 Module 3 (Cyber Physical System Models and Behaviours) 9 hours
3.1 Synchronous and Asynchronous Models 1hours
3.2 Reactive Components 1 hour
3.3 Components Properties 1 hour
3.4 Synchronous Designs and Circuits 1 hour
3.5 Asynchronous Processes and operations 1 hour
3.6 Design Primitives in Asynchronous Process 1 hour
3.7 Coordination Protocols in Asynchronous Process 1 hour
3.8 Leader Election 1 hour
3.9 Reliable Transmission. 1 hour
4 Module 4 (Security and Privacy in Cyber Physical Systems) 8 hours
4.1 Introduction to CPS securities, , 1 hour
4.2 Security and privacy issues in CPS 1 hour
4.3 Basic techniques in CPS security 1 hour
4.4 Cyber security requirements 1 hour
4.5 Attack model and Counter measures 2 hours
4.6 Case study 2 hours
5 Module 5 (CPS Application) 7 hours

5.1 Tools and programming frameworks for CPS 1 hours
5.2 Medical CPS 2 hours
5.3 Energy CPS 2 hours
5.4 CPS in Smart cities 2 hours
Reference Books
1. E. A. Lee and S. A. Seshia, “Introduction to Embedded Systems: A Cyber-Physical
Systems Approach”, 2011
2. R. Alur, “Principles of Cyber-Physical Systems,” MIT Press, 2015
3. Raj Rajkumar, Dionisio de Niz and Mark Klein, “Cyber-Physical Systems”, Addison-
Wesley, 2017
4. Fei Hu, “Cyber-Physical Systems”, CRC Press 2013
5. Houbing Song, Glenn A.Fink, Sabina Jesche, “Security and Privacy in Cyber-Physical
Systems: Foundations, Principles and Solutions”, IEEE Press.
6. Houbing song, Danda B Rawat, Sabina Jeschke, Christian Brecher, “Cyber Physical
Systems Foundations, Principles and Applications”, Elsevier, 2017
221ECS046 ADAPTIVE SIGNAL CATEGORY L T P CREDIT
PROCESSING PROGRAM 3 0 0 3
ELECTIVE 1
Preamble:
This course introduces the basic ideas of adaptation in the engineering sense and to set
adaptive signal processing into the general signal processing context. It enables the learners
to visualize the overall process of adaptation in geometrical terms. This course also
familiarize the basic methods of searching the performance surface and adaptive algorithms
and the applications of adaptive signal processing in several areas.
Basic knowledge in Digital Signal Processing is a prerequisite to learn the course
Course Outcomes:
CO 1 Explain basic meaning of adaptation in the engineering sense
CO 2 Think of the overall process of adaptation in geometrical terms
CO 3 Analyze and develop adaptive signal processing problems and algorithms.
CO 4 Set adaptive signal processing into the general signal processing context.

course.

bachelor program
aspects
CO 1
CO 2
CO 3
CO 4
Assessment Pattern
Apply 50-80%
Analyse 20-40%
Mark distribution
Marks
100 40 60 2.5 hours




1. Compare and contrast open loop and closed loop adaptivesystems
2 Explain in detail about adaptive linear combiner system with the help of block diagram

1. Discuss the geometrical significance of eigen vectors and eigen values
2. Derive an expression for performance function, Gradient and Minimum Mean Square Error

1. Explain LMS algorithm and convergence of weight vector in LMS algorithm
2. With relevant equations and diagrams explain the principle of Gradient Search method
3. Explain Gradient Search by Newton’s method
4 Explain misadjustment
5 Derive an expression for variance of gradient estimate

1. Explain in detail with example, how adaptive modelling can be applied to the multipath
problem in communication channel
2. Discuss adaptive model in geophysical exploration
.
QP CODE:
Reg No: _______________
Name: _________________ PAGES : 4
Course Name: Adaptive Signal Processing
PART A
1. Describe multiple input and single input adaptive linear combiner

2. Give an outline on the steepest descent method for gradient search
3. Discuss the terms performance penalty and excess MSE Describe the procedure
for estimating the gradient component by derivative measurement method
4. Discuss the terms performance penalty and excess MSE

5. Describe adaptive Modeling in Geophysical Exploration
(5x5=25
)
Part B
6. (a) Derive expression for performance function, gradient and minimum mean (7)
square error for an adaptive linear combiner
7. (a) Discuss the geometrical significance of eigen values and eigen vectors (4)
(b) Obtain the solution for simple gradient search algorithm (3)
8. (a) Describe the procedure for estimating the gradient component by derivative (4)
measurement method
b) Explain variance of the gradient estimate (3)
9. (a) Explain in detail about LMS algorithm and derive the equation. Discuss about (7)
convergence of weight vector
10. (a) Explain how adaptive modeling can be applied to the multipath problem in a (7)
communication channel
11. (a) Discuss about LMS Newton algorithm (4)

(b) Derive normal form of input correlation matrix (3)
12. (a) Discuss how the Z transform relates to the frequency response of an adaptive (7)
system
Syllabus
Module 1: (Adaptive Systems)
Definitions And Characteristics, Open Loop and Closed Loop Adaptation, Adaptive Linear ,
combiner, Performance Function, Gradient and MSE, Alternative Expression for Gradient
Module 2 Theory of Adaptation with stationary signals
Input Correlation Matrix, Eigen Values And Eigen Vectors Of Input Correlation Matrix,
Searching The Performance Function: Basic idea of Gradient Search, Simple Gradient Search
algorithm and its solution Stability and Rate of Convergence, Learning Curve, Newtons
Method, Steepest Descent Method, Comparison
Module 3 Gradient Estimation and its effects on adaptation
Gradient Component Estimation By Derivative Measurement, Performance Penalty, Variance

Of Gradient Estimate, Effects on Weight Vector Solution, Excess Mean Square Error
Misadjustments, Total Misadjustments and Other Practical Consideration.
Module 4: Adaptive algorithms, Structures, Z Transform
LMS Algorithm: derivation, convergence of weight vector, learning Curve, Noise vector in
weight vector solution, Misadjustment LMS Newton, Sequential Regression, Recursive Least
Squares, Z-Transform in Adaptive Signal Processing
Module 5: Application of adaptive signal processing

Adaptive Lattice Predictor, Adaptive Modeling : General description, Adaptive Modeling of
Multipath Communication Channel Adaptive, Modeling in Geophysical Exploration
Course Plan

(40 Hours )
1 Module 1 Adaptive Systems (7 hrs)
1.1 Definitions And Characteristics, Open Loop and Closed Loop Adaptation 2
1.2 Adaptive Linear Combiner 2
1.3 Performance Function, Gradient and MSE 2
1.4 Alternative Expression for Gradient 1
2 Module 2 Theory Of Adaptation With Stationary Signals (7 hrs)
2.1 Input Correlation Matrix, Eigen Values And Eigen Vectors Of Input Correlation Matrix 2
2.2 Searching The Performance Function: 3
Basic idea of Gradient Search, Simple Gradient Search algorithm and its solution
Stability and Rate of Convergence, Learning Curve
2.3 Newtons Method, Steepest Descent Method, Comparison 2

3 Module 3 Gradient Estimation and its Effects on Adaptation (7 hrs)
3.1 Gradient Component Estimation By Derivative Measurement, Performance Penalty 2
3.2 Variance Of Gradient Estimate, Effects on Weight Vector Solution 2
3.3 Excess Mean Square Error 2
3.4 Misadjustments, Total Misadjustments and Other Practical Consideration 1
4 Module 4 Adaptive Algorithms and Z Transform in ASP (11 hrs)
4.1 LMS Algorithm:
4.1.1 derivation, convergence of weight vector, learning Curve, 3
4.1.2 Noise vector in weight vector solution, Misadjustment 2
4.2 LMS Newton, Sequential Regression, Recursive Least Squares 3
4.3 Z-Transform in Adaptive Signal Processing 3
5 Module 5 Structures and Application of ASP (8 hrs)
5.1 Adaptive Lattice Predictor 2
5.2 Adaptive Modeling : General description 2
5.3 Adaptive Modeling of Multipath Communication Channel 2
5.4 Adaptive Modeling in Geophysical Exploration 2
Reference Books
1 Adaptive Signal Processing- Widrow and Stearns, Pearson
2 Statistical and Adaptive Signal Processing-Monalokis,Ingle and Kogon – Artech House
INC 2005
3 Adaptive Filter theory- 4th Edition Simon Haykin –Prentice Hall
4 Adaptive Filters- A H Sayed- John Wiley
5 Adaptive Filtering Primer with MATLAB- A Poularikas, Z M Ramadan, Taylor and
Francis Publications
221ECS050 CONVEX OPTIMIZATION CATEGORY L T P CREDIT
PROGRAM 3 0 0 3
ELECTIVE 1
Preamble:
The course aims to give students the tools and techniques to recognize and solve
convex optimization problems that arise in scientific and engineering
applications, presenting the basic theory, and concentrating on modeling aspects
and results that are useful in applications. Topics to be covered are the basic
theory of optimization, convex sets, convex functions, convex analysis, least-
squares, linear and quadratic programs, semidefinite programming, minimax,
optimality conditions and the duality theorem. Methods for solving optimization
problems including steepest descent, conjugate gradient, and interior point
methods are discussed in details.
Prerequisite: Basics of Vector calculus and Linear algebra.
Course Outcomes:
CO 1 Formulate and solve engineering problems via convex optimization methods

(Cognitive Knowledge Level : Apply)
CO 2 Recognize and characterize convex functions and sets (Cognitive Knowledge Level
: Apply)
CO 3 Demonstrate an understanding of the fundamental concepts of convex optimization

such as linear, quadratic, geometric and semi-definite programs, and duality theory.
(Cognitive Knowledge Level : Apply)
CO 4 Formulate the dual problem of some general optimization types and assess their
duality gap using concepts of strong and weak duality (Cognitive Knowledge Level
: Apply).
CO 5 Apply algorithms to solve standard unconstrained and constrained optimization

problems (Cognitive Knowledge Level : Apply).

course.
bachelor program
aspects
3 PO 1 PO 2 PO 3 PO 4 PO 5 PO 6 PO 7
CO 1
CO 2
CO 3
CO 4
CO 5
Assessment Pattern
Apply 50-80%
Analyse 20- 40%
Evaluate
Create
Mark distribution
Marks
100 40 60 2.5 hours




1. Suppose that there are m basic nutrients. A healthy diet needs bj units of jth nutrient per
day. There are n different food items available, with one unit of item i containing aji units
of nutrient j. Price of food item i is ci per unit. Formulate this as an optmization problem to
minimize the cost of food per day, while keeping the diet healthy.
2. In portfolio optimization, we seek the best way to invest some capital in a set of n assets.
The variable xi represents the investment in the ith asset, so the vector x ∈ Rn describes the
overall portfolio allocation across the set of assets. The constraints might represent a limit
on the budget (i.e., a limit on the total amount to be invested), the requirement that
investments are nonnegative (assuming short positions are not allowed), and a minimum
acceptable value of expected return for the whole portfolio. The objective or cost function
might be a measure of the overall risk or variance of the portfolio return. Formulate this as
an optimization problem that corresponds to choosing a portfolio allocation that minimizes
risk, among all possible allocations that meet the firm requirements.
3. Consider the task of choosing the width and length of each device in an electronic circuit.
Here the variables represent the widths and lengths of the devices. The constraints
represent a variety of engineering requirements, such as limits on the device sizes imposed
by the manufacturing process, timing requirements that ensure that the circuit can operate
reliably at a specified speed, and a limit on the total area of the circuit. A common
objective in a device sizing problem is the total power consumed by the circuit. The
optimization problem is to find the device sizes that satisfy the design requirements (on
manufacturability, timing, and area) and are most power efficient. Formulate this as an
optmization problem.
4. In data fitting, the task is to find a model, from a family of potential models, that best fits
some observed data and prior information. Here the variables are the parameters in the
model, and the constraints can represent prior information or required limits on the
parameters (such as nonnegativity). The objective function might be a measure of misfit or
prediction error between the observed data and the values predicted by the model, or a
statistical measure of the unlikeliness or implausibility of the parameter values. The
optimization problem is to find the model parameter values that are consistent with the
prior information, and give the smallest misfit or prediction error with the observed data.
Formulate this as an optmization problem.

1. Given an example of a closed set in R2 whose convex hull is not closed.
2. Show that if Si ⊆ Rn, i ∈ I is a collection of convex sets, then their intersection is also convex.
3. Let A ∈ Rm×n. Show that if S ⊆ Rn is convex then so is A(S) = {Ax : x ∈ S}, called the image of S
under A.
4. Give an example of a strictly convex function that does not attain its infimum.
5. Show that a set is convex if and only if its intersection with any line is convex. Show that a set is
affine if and only if its intersection with any line is affine.

1. Give an explicit solution to the following Linear Program.
2. Provide necessary and sufficient conditions under which a quadratic optimization problem
be written as a linear least squares problem.
3. State the first- and second-order conditions for optimality for Linear least squares and
Quadratic Optimization.

1. Express the dual problem of
with c not equal 0, in terms of the conjugate f. Explain why the problem you give is convex.
2. Find the dual function of the

Linear Program
3. For the following optimization problem, derive the KKT conditions. Find all solutions
that satisfy the KKT conditions. Which pair corresponds to the optimum?

1. Argue that the following unconstrained optimization problem is convex and formulate it as
second-order cone program (SOCP). Here x in Rn is the optimization variable, A in Rm x
n and b in Rm.
2. Define direction of descent, steepest descent direction, and the Newton direction.
3. Run the gradient descent line search method on the following function starting at the point
x = (2 , 2) and using only step sizes 1/2 for 3 steps.
QP CODE:
Reg No: _______________
Name: _________________ PAGES : 4
Course Name: CONVEX OPTIMIZATION
PART A
1. Show that a polyhedron {x ∈ Rn : Ax ≤ b}, for some A ∈ R m×n, b ∈ Rm, is both

convex and closed.
2. Prove that the maximum of a convex function over a bounded polyhedron must
occur at one of the vertices.
3. Consider the
optimization problem
Make a sketch of the feasible set for the objective functions f0(x1, x2) = x1 + x2.
4. Derive the dual of
What are the primal optimal value and the dual optimal value? What is the duality
gap?
5. Explain how to find a steepest descent direction in the ℓ2-norm, and give a simple
interpretation.
(5x5=25)
Part B
6. Let C ⊆ Rn be a convex set, with x1, . . . , xk ∈ C, and let θ1, . . . , θk ∈ R (7)
satisfy θi ≥ 0, θ1 + · · · + θk = 1. Show that θ1x1 + · · · + θkxk ∈ C.
7. Suppose f : R --> R is convex, and a, b ∈ dom f with a < b. Show that (7)
for all x ∈ [a, b].
8. Prove that x⋆ = (1, 1/2,−1) is optimal for the optimization problem (7)
9. (7)
where α is an integer between 0 and n. What happens if α is not an integer

(but satisfies 0 ≤ α ≤ n)?
10. (7)
11. (7)
11. Give the feasible set, the optimal value, and the optimal solution.
12. Derive the Lagrange dual function.
13. State the dual problem, and verify that it is a concave maximization
problem. Find the dual optimal value and dual optimal solution.
12. Consider a measurement model y = Ax + n, that is, observations y are given (7)
by the affine transformation of the input x in Rn with A in Rm x n plus additive
noise n. Given y, x is determined by solving the following problem
Find the Hessian of the objective function fo(x) with respect to x.

Syllabus
Module 1(Introduction to Convex optimization and Convex Sets)
General form of optimization problem, the basic convex optimization problem formulation. Important
classes of convex optimization, generalized problem, example: semidefinite programming and its
application in portfolio optimization. Convex sets: Basic types of convex sets and their formal
expressions, Operations that preserve convexity, Supporting hyperplane theorem and proof,
Separating hyperplane theorems and proofs.
Module 2(Convex Functions)
Definition and basic properties of convex functions, First and second order convexity conditions,
Sublevel sets and epigraph, Operations that preserve convexity, Quasiconvex functions and first and
second order quasiconvexity conditions, Operations that preserve quasiconvexity, Strongly convex
functions, First and second order conditions for strong convexity, Convexity with respect to
generalized inequialities.
Module 3(Convex optimization problems)
Optimization problem definition and examples, Optimality condition for differentiable functions,
Linear programming, Quadratic programming, Geometric programming, Semi-definite programming.
Module 4(Duality and Optimality conditions)
Lagrangian and Lagrange dual function, Lagrange dual problem, Properties, weak and strong duality,
Interpretation of dual variables, duality, Optimality conditions, Karush-Kuhn-Tucker (KKT),
necessity and sufficiency, Sub-gradients for non-smooth functions, Example: Water-filling and
reverse water-filling, Example: Compress sensing (using sub-gradient).
Module 5(Methods and Algorithms )
Unconstrained - Gradient descent, steepest descent, Newton method, With equality constraints -
Newton methods with equality constraints, With equality constraints - Newton methods with equality
constraints, With inequality constraints - Barrier interior point method, Barrier interior point method,
Primal-dual interior point methods.
Course Plan
No Topic No. of
Lectures (
38)
1 Module 1 (Introduction to Convex optimization and Convex Sets)
1.1 General form of optimization problem, the basic convex optimization 1

problem formulation
1.2 Important classes of convex optimization, generalized problem, example: 1
semidefinite programming and its application in portfolio optimization
1.3 Convex sets: Basic types of convex sets and their formal expressions 1
1.4 Operations that preserve convexity 1
1.5 Supporting hyperplane theorem and proof 1
1.6 Supporting hyperplane theorem and proof 1
1.7 Separating hyperplane theorems and proofs 1
1.8 Separating hyperplane theorems and proofs 1
2 Module 2 (Convex Functions)
2.1 Definition and basic properties of convex functions 1
2.2 First and second order convexity conditions 1
2.3 Sublevel sets and epigraph 1
2.4 Operations that preserve convexity 1
2.5 Quasiconvex functions and first and second order quasiconvexity 1

conditions
2.6 Operations that preserve quasiconvexity 1
2.7 Strongly convex functions 1
2.8 First and second order conditions for strong convexity 1
2.9 Convexity with respect to generalized inequialities 1
3 Module 3 (Convex optimization problems)
3.1 Optimization problem definition and examples 1
3.2 Optimality condition for differentiable functions 1
3.3 Linear programming 1

3.4 Quadratic programming 1
3.5 Geometric programming 1
3.6 Semi-definite programming 1
4 Module 4 (Duality and Optimality conditions)
4.1 Lagrangian and Lagrange dual function 1
4.2 Lagrange dual problem, Properties, weak and strong duality 1
4.3 Interpretation of dual variables, duality 1
4.4 Optimality conditions, Karush-Kuhn-Tucker (KKT), necessity and 1

sufficiency
4.5 Sub-gradients for non-smooth functions 1
4.6 Example: Water-filling and reverse water-filling 1
4.7 Example: Compress sensing (using sub-gradient) 1
5 Module 5 (Methods and Algorithms)
5.1 Unconstrained - Gradient descent, steepest descent 1
5.2 Newton method 1
5.3 With equality constraints - Newton methods with equality constraints 1
5.4 With equality constraints - Newton methods with equality constraints 1
5.5 With inequality constraints - Barrier interior point method 1
5.6 Barrier interior point method 1
5.7 Primal-dual interior point methods 1
5.8 Primal-dual interior point methods 1
Reference Books
1.Boyd and Vandenberghe, Convex Optimization, Cambridge University Press, 2004
2.Bertsekas, Nedic, and Ozdaglar, Convex Analysis and Optimization
3.Convex Analysis By R.T.Rockafellar,Pub:Princeton
4.Convex Analysis and Nonline,ptimization, Borwein & Lewis

221ECS047 ROBOTICS & AUTOMATION CATEGORY L T P CREDIT
PROGRAM 3 0 0 3
ELECTIVE 1
Preamble:
Robotics and industrial automation is changing the face of production. Manufacturers around
the globe are implementing some form of automation to become more efficient, and safe and
ultimately increase revenues. This course helps the student with the basic idea of Robots and
industrial automation. This course introduces the industrial automation process, basic robot
anatomy and familiarise students to various sensors, end effectors and actuators, concepts of
kinematics and inverse kinematics. This course helps students to develop robotic based
solutions to real world problems.
Course Outcomes: After the completion of the course the student will be able to
CO 1 Analyse principles, strategies and elements of automation process (Cognitive

Knowledge Level :Analyse)
CO 2 Explain the terminologies of robotics and the working mechanism of grippers

(Cognitive Knowledge Level :Apply)
CO 3 Illustrate the concept of robot kinematics (Cognitive Knowledge Level :Apply)
CO 4 Choose the suitable sensors, actuators and control for robot design (Cognitive Knowledge
Level :Apply)
CO 5 Explain various actuators and drive systems for robot applications (Cognitive
Knowledge Level :Apply)
CO 6 Design, Develop, Implement and Present innovative ideas by applying principles and
techniques of Robotics and Automation (Cognitive Knowledge Level: Create)

course.
PO1:An ability to independently carry out research/investigation and developmentwork in

PO2:An ability to communicate effectively, write and present technical reports on complex
PO3:An ability to demonstrate a degree of mastery over the area as per the specialization of the
program. The mastery should be at a level higher than the requirements in the appropriate bachelor
program
PO4:An ability to apply stream knowledge to design or develop solutions for real world problems by
PO5:An ability to identify, select and apply appropriate techniques, resources and state-of-the-art
PO6:An ability to engage in life-long learning for the design and development related to the stream
related problems taking into consideration sustainability, societal, ethical and environmental aspects
PO7:An ability to develop cognitive load management skills related to project management and
CO 1
CO 2
CO 3
CO 4
CO5
CO6
Assessment Pattern
Apply 50-80%
Analyse 20-40%
Evaluate
Create
Mark distribution
Total CIE ESE ESE
Marks Duration
100 40 60 2.5 hours

Evaluation shall only be based on application, analysis or design-based questions (for both

ii.Course based task / Seminar/ Data collection and interpretation : 15 marks

marks for each question. Students should answer all questions. Part B will contain 7 questions
(such questions shall be useful in the testing of overall achievement and maturity of the
students in a course, through long answer questions relating to theoretical/practical
knowledge, derivations, problem solving and quantitative evaluation), with minimum one
question from each module of which student should answer any five. Each question can carry
7 marks
For example, if the average end semester mark % for a core course is 40, then the maximum eligible
1. How automation helps in industry?

2 Analyse Continuous vs Discrete Control systems.
3. Discuss Automation principles and strategies.
1. How precision and accuracy help a robot in industrial automation?
2. Find out a suitable application for magnetic type gripper in the industry.
3. What is the importance of a biologically inspired robot?
1. What do you mean by forward kinematics?
2. Explain the inverse kinematics of robots.
1. Which type of sensor is useful to measure rpm and direction of the flywheel
2. Elucidate various sensors used in robots
3. Discuss various stages in image processing techniques
1. How is actuation possible in robots?
2. Design a system to control the speed and direction of a DC motor
3. An industrial system needs to drive a wheel that is 3 m away from the motor. Which type
of drive system is useful and why?
QP CODE:
Reg No: _______________
Name: _________________
PAGES : 4
Course Name: ROBOTICS & AUTOMATION
PART A
1. Explain the role of automation in industry.
2. Discuss briefly about Biologically Inspired Robot Design.
3. Analyse the direction control of motors using H-bridge.
4. What is a homogeneous transformation matrix? Explain its significance.

Explain each element of the matrix. Explain the vectors n, o and a.
5. Distinguish between Shape Memory Alloy (SMA), Electro Active Polymers

(EAP) (5x5=25)
Part B
6. Explain the robot anatomy with a neat diagram. Explain the major (7)
components of a robot.
7. A) Explain the transformations involved in the cylindrical coordinate (7)

system. Derive the transformation matrix for the same.
B) A point P (7, 3, 1)T is attached to a moving frame Fnoa and is subjected
to the following transformations. Find the coordinates of the point relative
to the reference frame at the conclusion of transformations.
1. Rotation of 90 degrees about the z-axis
2. Followed by a rotation of 90 degrees about the y-axis
3. Followed by a translation of [4, -3,7] in x, y, z directions.
8. What is the role of sensors in robots? Explain their characteristics. Discuss (7)
various force and torque sensors in robotics.
9. a. A) Justify the statement: “Actuators are the muscles of robots.” (2)
b. Illustrate the difference between linear and rotary actuation mechanisms (5)
using hydraulic drives.
10. a. Explain the various image processing techniques in robotics. (4)
b. Explain an edge detection technique. (3)
11. Compare and contrast the working principle of hydraulic, pneumatic and (7)
electrical actuators
12 Explain the following types of sensors (7)

i. Position and displacement sensors
ii. Optical sensors
iii. Force-torque sensors
Syllabus
Module 1 (Introduction to automation)
Definition of Automation, Understanding production systems, Automation in production

systems, Automation principles and strategies, Basic elements of the automated system,
Advanced Automation Functions - Levels of automation, Advanced Automation Functions -
Levels of automation, Industrial Control systems, Process industries vs Discrete
manufacturing Industries, Continuous vs Discrete Control, Basics of Different systems for
Industrial Automation: PLC, HMI, SCADA, DCS
Module 2 (Introduction, Definitions- Robots, Robotics)

Introduction to Robotics, Evolution of Robots and Robotics. Robot Anatomy (Robotic Arm):
Links, Joints and joint Notation Scheme, Degrees of Freedom, Arm Configuration, Wrist
Configuration, End effector, Robot Characteristics-Repeatability, Precision and Accuracy,
End-effectors, Grippers - Mechanical Grippers, Pneumatic and Hydraulic Grippers, Magnetic
Grippers, Vacuum Grippers, Applications of Robots, Introduction to bio-inspired robots:
Biologically Inspired Robot Design, Biologically Inspired Robot Control, Biologically
inspired actuation and sensing.
Module 3 (Introduction to Robot Kinematics)

World frame, end-effectors frame, Position and orientation of objects, Translation and
Rotation, Rotation matrix, Arm equation, Forward kinematics, Homogeneous Transformation
Matrix, Euler angles, Denavit-Hartenberg (DH) representation, Inverse kinematics
Module 4 (Robotic Sensors)

Different types and applications of sensors in robotics, Position and displacement sensors,
Optical sensors, Force-torque sensors, Tachometer’s Direction of rotation, Velocity and
acceleration measurements, Characteristics of sensors, Robotic vision systems, Image
processing techniques: Basic Concepts, Fourier transform and Frequency content of a signal
and an Image. Resolution and Quantization. Image Processing techniques: Histogram,
Thresholding, Noise reduction using image averaging, Edge Detection.
Module 5 (Robot Actuators and drive systems)

Characteristics of Actuating systems, Comparison of Actuating systems, Actuators-Hydraulic
and Pneumatic, Electrical actuators: DC motors, AC motors, Stepper motors,Speed control
using PWM and direction control using H-bridge, Shape Memory Alloy (SMA), Electro
Active Polymers (EAP), Gears, Gear train, Belts and types of belts
Course Plan
No Topic No. of
Lectur
es
1 Introduction to automation
1.1 Definition of Automation, 1
1.2 Understanding production systems, Automation in production systems. 1
1.3 Automation principles and strategies. 1
1.4 Basic elements of the automated system 1
1.5 Advanced Automation Functions - Levels of automation 1

1.6 Industrial Control systems, Process industries vs Discrete manufacturing 1
Industries, Continuous vs Discrete Control.
1.7 Basics of Different systems for Industrial Automation: PLC, HMI, 1
SCADA, DCS
2 Introduction, Definitions- Robots, Robotics
2.1 Introduction to Robotics, Evolution of Robots and Robotics. Robot 2
Anatomy (Robotic Arm): Links, Joints and joint Notation Scheme,
Degrees of Freedom
2.2 Arm Configuration, Wrist Configuration, End effector, Robot 2
Characteristics-Repeatability, Precision and Accuracy
2.3 End-effectors, Grippers - Mechanical Grippers, Pneumatic and 2

Hydraulic Grippers, Magnetic Grippers, Vacuum Grippers
2.4 Applications of Robots 1
2.5 Introduction to bio-inspired robots: Biologically Inspired Robot Design, 2
Biologically Inspired Robot Control, Biologically inspired actuation and
sensing.
3 Introduction to Robot Kinematics
3.1 World frame, end-effectors frame, Position and orientation of objects 2
3.2 Translation and Rotation, Rotation matrix, Arm equation 2

3.3 Forward kinematics, Homogeneous Transformation Matrix, Euler angles
2
3.4 Denavit-Hartenberg (DH) representation, Inverse kinematics 2
4 Robotic Sensors
4.1 Different types and applications of sensors in robotics, Position and
displacement sensors, Optical sensors, 1
Force-torque sensors, Tachometer’s Direction of rotation 1
Velocity and acceleration measurements, Characteristics of sensors 1
4.2 Robotic vision systems, Image processing techniques: Basic Concepts
1
4.3 Fourier transform and Frequency content of a signal and an Image. 2
Resolution and Quantization.
4.4 Image Processing techniques: Histogram, Thresholding, Noise reduction 2
using image averaging, Edge Detection.
5 Robot Actuators and drive systems
5.1 Characteristics of Actuating systems, Comparison of Actuating systems
2
5.2 Actuators-Hydraulic and Pneumatic, Electrical actuators: DC motors,
AC motors, Stepper motors,
2
5.3 Speed control using PWM and direction control using H-bridge
1
5.4 Shape Memory Alloy (SMA), Electro Active Polymers (EAP)
1
5.5 Gears, Gear train, Belts and types of belts
1
Reference Books
1. Mikell P. Groover, Automation, Production systems, and Computer-Integrated

Manufacturing. PHI (2011)
2. R.K Mittal, I.J Nagrath, “Robotics & Control” Tata McGraw-Hill Education Private Ltd.
(2003)
3. Yunhui Liu and Dong Sun, “Biologically Inspired Robotics”, CRC Press (2012)
4. Saeed B Niku, “Introduction to robotics: analysis, control, applications”, John Wiley &
Sons, (2020)
5. Groover, Mikell P., Mitchell Weiss, and Roger N. Nagel. “Industrial robotics: technology,
programming and application”, McGraw-Hill Higher Education, (1986)
6. Fu, K. S., and RC Gonzales Robotics. "Control, sensing, Vision, and intelligence”, Mc.
Graw Hill (1989)
7. Ashitava Ghoshal, “Robotics: fundamental concepts and analysis”, Oxford university
press, (2006)
8. W. Bolton, “Mechatronics: Electronic Control systems in Mechanical and Electrical
Engineering.” Pearson Education.
9. Jisu Elsa Jacob, Manjunath N, “Robotics simplified”, BPB Publications (2022)
PROGRAM
221ECS048 CYBER FORENSICS 3 0 0 3
ELECTIVE 1
Preamble: The course on Cyber Forensics aims at exploring the basics of Cyber Forensics
and Cyber security, the forensic investigation process and principles and the different types of
cybercrimes and threats. The course also focuses on the forensic analysis of File systems, the
Network, the Windows and Linux Operating systems. The course gives a fundamental
understanding of the forensics analysis tools and a deep understanding of Anti forensics
practices and methods.
CO# Course Outcomes

Analyse the concepts in cyber forensics, forensics Investigation Process and the
CO1 usage of Cyber Forensics Tools in investigations (Cognitive Knowledge Level:
Analyse)
Infer the attribute definitions of file systems structure (Cognitive Knowledge Level:
CO2
Apply)
Utilize the methodologies used in memory analysis and network analysis for
CO3
detection of artifacts (Cognitive Knowledge Level: Apply)
Examine the significance of IT Act 2000 and relate it with respect to current cyber
CO4
security scenarios. (Cognitive Knowledge Level: Analyze)
Use conventional practices and techniques for antiforensics detection (Cognitive
CO5
Review forensic investigation procedures by applying cyber forensic methodologies
CO6
and tools (Cognitive Knowledge Level: Evaluate)

course.

the program. The mastery should be at a level higher than the requirements in the appropriate
bachelor program
CO 1
CO 2
CO 3
CO 4
CO 5
CO 6
Assessment Pattern
Apply 60%
Analyse 20%
Evaluate 20%
Create
Mark distribution
Total
CIE ESE ESE Duration
Marks
100 40 60 2.5 hours



Part A will contain 5 numerical/short answer questions with 1 question from each module,
having 5 marks for each question. Students should answer all questions. Part B will contain 7
questions (such questions shall be useful in the testing of overall achievement and maturity of
the students in a course, through long answer questions relating to theoretical/practical
7 marks
Note: The marks obtained for the ESE for an elective course shall not exceed 20% over the
average ESE mark % for the core courses. ESE marks awarded to a student for each elective
course shall be normalized accordingly.
For example if the average end semester mark % for a core course is 40, then the maximum
eligible mark % for an elective course is 40+20 = 60 %.
Syllabus
MODULE-1 (CYBER FORENSICS)
Cyber Forensics: Cyber Technology- Technological Aspects of Cyber Forensics-

Governance Aspects of Cyber Forensics- Judicial Aspects of Cyber Forensics- Legal
Perspective of Cyber Forensic investigations.
Computer Forensic Investigations: -Preparing for computer investigations, understanding

Public and private investigations, Forensics Investigation Principles - Forensic Protocol for
Evidence Acquisition - Digital Forensics Standards and Guidelines - Digital Evidence – Data
Acquisition - storage formats for digital evidence, determining the best acquisition method.
MODULE-2: (CYBER FORENSICS TOOLS AND TYPES OF FORENSICS)
Cyber Forensics Tools-Computer Forensics software and hardware tools -Open Source and
Proprietory -Challenges in Cyber Forensics, Skills Required to Become a Cyber Forensic
Expert- Physical Requirements of a Cyber forensics Lab, Types of Cyber forensics.
File System Forensics-Working with windows and DOS systems- file systems, exploring
Microsoft file structures, examining FAT and NTFS disks, whole disk encryption, the
windows registry, Microsoft and MS-DOS start-up tasks, Examining UNIX and LINUX disk
structures and boot processes.
MODULE-3 (OS AND NETWORK FORENSICS)
Windows and Linux Forensics:
Windows Forensics-Live Response: Data Collection- Introduction, Locard’s Exchange

Principle, Order of Volatility - Volatile and Non-Volatile Data Live-Response
Methodologies: Data Analysis, Windows Memory Analysis, Rootkits and Rootkit detection.
Linux Forensics: Live Response Data Collection- Data Analysis- Log Analysis, Keyword
Searches, The Hacking Top Ten, Reconnaissance Tools.
Network Forensics: The OSI Model, Forensic Footprints, Seizure of Networking Devices,
Network Forensic Artifacts, ICMP Attacks, Drive-By Downloads, Network Forensic
Analysis Tools, Case Study: Wireshark. Web Attack Forensics: OWASP Top 10, Web Attack
Tests, Penetration Testing.
MODULE-4 (CYBER SECURITY)
Cyber Security: Cybercrimes, Types of Cybercrimes –Cyber Security Steps taken to protect
ICT and prevent Misuse of Internet- IT Act 2000- Social Cyber Media.
MODULE-5: (ANTI-FORENSICS):
Anti-forensic Practices - Data Wiping and Shredding- Data Remanence,

Degaussing, Case Study: USB Oblivion, Eraser - Trail Obfuscation: Spoofing, Data
Modification, Case Study: Timestamp – Encryption, Case Study: VeraCrypt, Data Hiding:
Steganography and Cryptography, Case Study: SilentEye, Anti-forensics Detection
Techniques, Case Study: Stegdetect
TEXT BOOKS
1. Nishesh Sharma “Cyber Forensics in India- A Legal Perspective”, Universal Law
Publishing, First Edition, March 2017
2. Bill Nelson, Amelia Philipps and Christopher Steuart, “Computer forensics- Guide to
computer forensics and investigations”, Course Technology Inc,3rd Edition,2009
3. Brian Carrier “File System Forensic Analysis, Addison-Wesley,1st Edition, 2005
4. Harlan Carvey, “Windows Forensics Windows Forensic Analysis DVD Toolkit”
O’Reilly,2nd Edition, 2018.
REFERENCES
1. Chris Pogue , Cory Altheide, Todd Haverkos, “Linux Forensics- Unix and Linux
Forensic Analysis DVD Toolkit”,Syngress, First Edition,2008
2. William Stallings “Network Security Essentials Applications and Standards” Pearson
Education, 4th Edition, 2011.
3. E. Maiwald, “ Fundamentals of Network Security”, McGraw-Hill, First Edition, 2004
4. Michael. E. Whitman, Herbet. J. Mattord,“Cyber Security Principles of Information

Security”,Course Technology Ptr ,4th Edition,2011.
5. William Stallings “Cryptography and Network Security”,Pearson, 5th Edition,2018.
6. Niranjan Reddy, “Practical Cyber Forensics: An Incident-Based Approach to Forensic
Investigations”, Apress, First Edition, 2019

1. Use a sample case scenario and apply the Forensics principles and protocols for its
evidence acquisition.
1. Outline the pros and cons of NTFS and FAT File systems. Also give the challenges
the Investigators would face in extracting evidence from these file systems.
2. Compute the file and RAM slack space of a FAT16 drive for placing a 5000
character file.
1. Use any memory forensics methodologies/tools to extract volatile and nonvolatile data
from a Windows based system.
2. Use web attacks test tools like netcraft to identify web application vulnerabilities of a
particular site say www.xyz.com

1. Cite the appropriate measures which can be used to protect ICT and to prevent Misuse
of Internet.
2. Discuss the impact of Section 66 in IT Act to General Public.

1. Use any anti-forensics practices /methods to destroy or conceal data in order to
prevent others from accessing it and write down the inferences from it.

1. Write down a report on an evidence analysis of a private case investigation by applying
forensic investigation procedures and by using cyber forensic methodologies and tools
QP CODE:
Reg No: _______________
Name: _________________ PAGES : 4
Course Name: CYBER FORENSICS
PART A
1. Distinguish between public and private investigations.
2. How file slack, RAM slack and drive slack be computed?

Why is it important to collect non-volatile information in forensic
3.
investigations?
4. What are the different tools used in Network Forensics?
5. How is data wiping done on the hard drive? (5x5=25)
Part B
6. Apply the principles of Digital Forensic Investigation for an internet abuse (7)
investigation and prepare a case report?
7. (a) Does Windows NT use FAT or NTFS. Explain (4)
(b) List the different Open Source Tools used in Image Acquisition. (3)
8. (a) What is Locard’s Exchange Principle? Explain with suitable examples (4)
(b) What is volatility? How is it determined? (3)
9. What is OWASP? What is the significance of web application (7)

vulnerabilities in the maintenance of a website?
10. Write down the impact of cybercrimes in society? How these crimes be (7)
prevented and showcase the mitigation steps used for the prevention of
Cybercrimes
11. (a) Why does data need Cryptography? (4)
(b) Differentiate a Cryptographer and a Crypter with suitable examples. (3)
12. (a) Trace out how encryption is done using Veracrypt (7)
content can be for 26 hrs. The audit course in third semester can have content for 30 hours)
No. of
No. Topic Lectures
( 40)
1 Module-1 (Cyber Forensics) (9 hrs)
1.1 Cyber Technology- Technological Aspects of Cyber Forensics 1
Governance Aspects of Cyber Forensics- Judicial Aspects of Cyber

1.2 1
Forensics
1.3 Legal Perspective of Cyber Forensic investigations 1
Preparing for computer investigations, understanding Public and

1.4 1
private investigations
Forensics Investigation Principles - Forensic Protocol for Evidence

1.5 1
Acquisition
1.6 Digital Forensics Standards and Guidelines 1
1.7 Digital Evidence and Data Acquisition 1
1.8 Storage formats for digital evidence 1
1.9 Determining the best acquisition method 1
2 Module-2 (Cyber Forensics Tools and Types of Forensics) (8 hrs)
2.1 Computer Forensics software and hardware tools -Open Source and 1
Proprietary
Challenges in Cyber Forensics, Skills Required to Become a Cyber
2.2 1
Forensic Expert
Physical Requirements of a Cyber forensics Lab,Types of Cyber

2.3 1
forensics
2.4 Working with windows and DOS systems- file systems 1
2.5 Exploring Microsoft file structures, examining FAT and NTFS disks 1
2.6 Whole disk encryption, Windows registry 1
2.7 Microsoft and MS-DOS start up tasks 1
2.8 Examining UNIX and LINUX disk structures and boot processes 1
3 Module-3 (Operating System Forensics) (11 hrs)
3.1 Windows Forensics-Live Response: Data Collection- Introduction 1
3.2 Locard’s Exchange Principle, Order of Volatility - Volatile and Non 1

Volatile Data
3.3 Linux Forensics:Live Response Data Collection- Data Analysis- Log 1

Analysis.
3.4 The Hacking Top Ten, Reconnaissance Tools- I 1
3.5 The Hacking Top Ten, Reconnaissance Tools- II 1
3.6 The OSI Model, Forensic Footprints 1
Seizure of Networking Devices, Network Forensic Artifacts, ICMP

3.7 1
Attacks
3.8 Drive-By Downloads, Network Forensic Analysis Tools 1
3.9 Case Study: Wireshark. 1
3.10 Web Attack Forensics: OWASP Top 10, Web Attack Tests 1
3.11 Penetration Testing 1
4 Module-4 (Cyber Security) (6 hrs)
4.1 Cybercrimes, Types of Cybercrimes 1
Cyber Security Steps taken to protect ICT and prevent Misuse of

4.2 1
Internet
4.3 IT Act 2000- Lecture I 1
4.4 IT Act 2000- Lecture II 1
4.5 Social Cyber Media- Lecture I 1
4.6 Social Cyber Media- Lecture II 1
5 Module – 5 (Anti-Forensics) (6 Hrs)
5.1 Anti-forensic Practices - Data Wiping and Shredding 1
5.2 Data Remanence, Degaussing 1
5.3 Trail Obfuscation: Spoofing, Data Modification 1
5.4 Role of Encryption in Forensics 1
5.5 Data Hiding: Steganography and Cryptography 1
5.6 Anti-forensics Detection Techniques 1

221ECS049 DATA ANALYTICS CATEGORY L T P CREDIT
PROGRAM 3 0 0 3
ELECTIVE 2
Preamble:
This course enables the students to understand the concepts of Data Analytics. It covers
Data and Relations, Correlation, Basic Data Analytics and visualization methods using R,
Finite State Machines, Dimensionality reductions, Feature extraction, Clustering,
Classification and Regression Techniques, and scalability through parallelization. It helps the
learners to develop applications for real time data analysis.
Course Outcomes:
CO 1 Identify data errors and dependencies among attributes by modelling them as sets &
relations. (Cognitive Knowledge Level: Apply)
CO 2 Apply statistical methods for evaluation hypothesis (Cognitive Knowledge Level:

Apply)
CO 3 Apply regression, classification, and clustering models on a given dataset (Cognitive

CO4 Apply correlation techniques to find the dependencies between the features.
CO5 Develop applications that uses the concepts in Data Analytics (Cognitive
Knowledge Level: Create)

course.

engineering and allied streams.
bachelor program.
problems by following the standards.
environmental aspects.
CO 1 ✔ ✔ ✔ ✔ ✔
CO 2 ✔ ✔ ✔ ✔ ✔
CO 3 ✔ ✔ ✔ ✔ ✔
CO 4 ✔ ✔ ✔ ✔ ✔
CO5 ✔ ✔ ✔ ✔ ✔ ✔ ✔
Assessment Pattern
Bloom’s Category End Semester

Examination
Apply 60-80%
Analyse 20-40%
Evaluate
Create
Assignments or course projects can be used for higher level assessment of course outcomes.
Mark distribution
Total CIE ESE ESE

Marks Duration
100 40 60 2.5 hours


students.
7 marks

1. How to model stochastic and deterministic errors. Explain with examples.
2. What are the ways in which various errors can be handled ?

1. What R commands would you use to remove null values from a data set.
2. Which function R can be used to a fit nonlinear line to the data.

1. Consider the data sets for two classes X1 = {(0,0)} and X2 = {(1,0), (0,1)}. Which
classification probabilities will a naive Bayes classifier produce for the feature vector (0,0)?
2. Explain SVM classifier with an example.

1. For the data set X = { (1,0), (2,0), (3,1) , (4,1), (5,1), (6,1), (7,0), (8,0) } compute chi-
square test statistic for 4 bins.
2. Explain the difference between correlation and causality.

1.Develop a small application for a manufacturing industry to maintain their works using the
concepts in data analytics.
2. Develop a small application for improving Transportation System using the concepts in
data analytics.
QP CODE:
Reg No:
Name: PAGES : 2
Course Name: DATA ANALYTICS
Max. Marks : 60 Duration: 2.5

Hours
PART A
1. Compute the output of (a) an asymmetric moving mean filter, q =3, (b) an
asymmetric moving median filter, q = 3, (c) an exponential filter, y0 = 0,
correction term n= 0.5 for the time series (0,0,0,1,0,0,0 ). Which filter result do
you like best?
2. Differentiate between Type I and Type II errors.
3. Construct the data tuples for an autoregressive forecasting model with a time
horizon of m = 2 for the time series x = (1,2,3,5,8).
4. Explain prototype based clustering.

5. Explain Rectified linear activation unit. (5x5=25)
Part B
6. Explain stochastic and deterministic errors with examples. Using 2-sigma rule(7)
and m-sigma rule how a value is classified as outliers?
7. Consider the data sets for two classes X1 = {(0,0)} and X2 = {(1,0), (0,1)}.(7)
Which classification probabilities will a naive Bayes classifier produce for the
feature vector (0,0)?
8. Illustrate the importance of visualizing data before analysis. (7)
9. What is feature scaling? Explain the feature scaling techniques. (7)
10 Justify how the computational complexity of the nearest neighbor is reduced by (7)
the LVQ approach.
11 For the data set X = { (1,0), (2,0), (3,1), (4,1), (5,1), (6,1), (7,0), (8,0) } (7)
compute chi-square test statistic for 4 bins.
12 Consider the two dimensional patterns (2, 1), (3, 5), (4, 3), (5, 6), (6, 7), (7, 8). (7)
Compute the principal component using PCA Algorithm. Use PCA Algorithm
to transform the pattern (2, 1) onto the Eigen vector.
Syllabus: Error Handling, Correlation, Models, Clustering, Data and Process Parallelization,
Batch processing frameworks.
Syllabus
Module Content Hours
1 Data and Relations - Data scales, Set and Matrix representations, 7

Relations, Similarity and dissimilarity measures, Sequence relations. Data
preprocessing - Error types, error handling, filtering, transformation,
merging.
Correlation - Linear, Causality, Chi-Square tests. Cross validation and

feature selection.
2 Basic Data Analytics Methods Using R - Descriptive Statistics, 9

Statistical methods for evaluation, Hypothesis Testing, ANOVA.
Visualization methods using R - Exploratory Data Analysis,

Visualizing single Variable, Examining Multiple Variables
3 Models- Finite state machines, Recurrent models, Autoregressive models, 6

Moving Average Models.
4 Clustering - Cluster partitions, Sequential clustering, Prototype based 7

clustering, Relational clustering, Cluster tendency assessment, Cluster
validity, Self organising map (SOP). Use Cases
Regression- Linear Regression, Logistic regression , Use Cases
5 Classification Methods- Naive Bayes classifier, Decision Trees, LDA, 11

SVM, Learning Vector Quantization.
Scalability through parallelization - Data parallelization, Process

parallelization, Scaling using feature engineering, Dimensionality
Reduction, Cascading, Feature reduction through spatial transforms. Global
Average Pooling, Data Augmentation,
Case Studies : ReLU nonlinearity, MLP, Convolutional Layer.

Course Plan
1 Data and Relations, Correlation
1.1 Data scales, Set and Matrix representations 1
1.2 Relations, Similarity and dissimilarity measures 1
1.3 Sequence relations. 1
1.4 Data pre-processing - Error types, error handling, Filtering 1
1.5 Transformation, merging 1
1.6 Correlation-Linear, Causality Chi-Square tests 1
1.7 Cross validation and feature selection 1
2 Basic Data Analytics Methods Using R
2.1 Introduction to R and GUI 1
2.2 Attributes and Data Types 1
2.3 Descriptive Statistics 1
2.4 Statistical methods for evaluation and Hypothesis 1
2.5 Hypothesis Testing- Difference of Means 1
2.6 Type –I, Type –II Errors, Problems 1
2.7 ANNOVA 1
2.8 Visualization – Single variable. 1
2.9 Examining multiple variables 1

3 Models
3.1 Finite state machines 1
3.2 Recurrent models 1
3.3 Autoregressive models 1
3.4 Autoregressive models 1
3.5 Moving Average Models- ARIMA 1
3.6 Moving Average Models- ARMA 1
4 Clustering
4.1 Cluster partitions 1
4.2 Sequential clustering, Prototype based clustering 1
4.3 Relational clustering, 1
4.4 Cluster tendency assessment, Cluster validity 1
4.5 Self-organising map (SOP). 1
4.6 Regression: Linear regression 1
4.7 Logistic Regression, Use cases 1
5 Classification
5.1 Naive Bayes classifier 1
5.2 LDA 1
5.3 SVM 1
5.4 Learning Vector Quantization 1

5.5 Scalability through parallelization 1
5.6 Data parallelization, Process parallelization 1
5.7 Scaling using feature engineering 1
5.8 Cascading, Feature reduction through spatial transforms 1
5.9 ReLU nonlinearity 1
5.10 Data Augmentation, MLP, Convolutional Layer 1
5.11 Global Average Pooling, Dimensionality Reduction 1
Reference Books
1. Thomas A. Runkler, “Data Analytics - Models and Algorithms for Intelligent Data
Analysis”, Springer 2012.
2. Stefanos Vrochidis, Benoit Huet, Edward Chang, IoannisKompatsiaris, “Big Data

Analysis for Large-Scale Multimedia Search”, Wiley 2019.
3. J. O. Moreira, Andre Carvalho, Tomas Horvath, “A General Introduction to Data

Analytics”, Wiley 2019.
4. “Data Science and Big data Analytics, Discovering, Analyzing, Visualizing and
Presenting Data” EMC Education Service.
5. Gouzhu Dong and Huan Liu, “ Feature Engineering For Machine Learning and Data
Analytics “ CRC Press..
221ECS007 PATTERN RECOGNITION CATEGORY L T P CREDIT
PROGRAM
3 0 0 3
ELECTIVE 2
Preamble: This course aims to impart the fundamentals of statistical pattern recognition
and neural network techniques. It introduces to the learner the various pattern recognition
algorithms, feature selection, classification, clustering and the use of neural networks in
feature extraction. This helps the learner to apply the algorithms in applications that works on
pattern recognition and machine intelligence.
Course Outcomes:After the completion of the course the student will be able to
CO 1 Apply probability and numerical methods in statistical pattern recognition.

CO 2 Apply statistical methods in feature selection. (Cognitive Knowledge Level: Apply)
CO 3 Apply linear algebra and statistical methods in parameter and non-parameter
estimation. (Cognitive Knowledge Level: Apply)
CO 4 Apply the technique of decision trees in pattern recognition. (Cognitive Knowledge
Level: Apply)
CO 5 Analyze the use of deep learning networks andartificial neural networks in pattern
recognition. (Cognitive Knowledge Level: Analyze)
CO 6 Design, Develop, Implement and Present innovative ideas in problem solving with
various pattern recognition techniques. (Cognitive Knowledge Level: Create)

course.

program
CO 1
CO 2
CO 3
CO 4
CO 5
CO 6
Assessment Pattern
Apply 70%-80%
Analyze 30%-40%
Evaluate
Create
Mark distribution
Marks
100 40 60 2.5 hours

Evaluation shall only be based on application, analysis or design-based questions (for both
ii.Course based task / Seminar/ Data collection and interpretation : 15 marks

students in a course, through long answer questions relating totheoretical/practical
7 marks
For example, if the average end semester mark % for a core course is 40, then the maximum eligible

1. Illustrate the design cycle of a pattern recognition system with the help of an example.
2. Suppose that we have three coloured bottles r (red), b(blue) and g(green).Box r
contains 3 apples, 4 oranges and 3 limes. Box B contains 1 apple, 1 orange and 0 limes and
box g contains 3 apples, 3 oranges and 4 limes. If a box is chosen at random with probability
p(g)=0.2, p(b)=0.2 and p(g)=0.6 and piece of fruit is removed from the box(with equal
probability of selecting items from the box), then what is the probability of selecting an
apple? If we observe that the selected fruit is in fact an orange, what is the probability that it
came from the green box?

1. Illustrate feature selection using t-Test with the help of an example.

1. Derive the fuzzy C spherical shells (FCSS) algorithm for the case that spherical
clusters are to be identified.

1. Illustrate decision tree with the help of an example. How does it enable pattern
classification?
2. Construct a decision tree using the following data.
Outlook Temp Humidity Windy Play Golf

Rainy Hot High False No
Rainy Hot High True No
Cloudy Hot High False Yes
Sunny Mild High False Yes
Sunny Cool Normal False Yes
Sunny Cool Normal True No
Cloudy Cool Normal True Yes
Rainy Mild High False No
Rainy Mild Normal False Yes
Sunny Mild Normal False Yes
Rainy Cool Normal True Yes
Cloudy Mild High True Yes
Cloudy Mild Normal False Yes
Sunny Hot High True No
1. How do artificial neural networks play a significant role in pattern recognition? Also
discuss about its parameter optimisation techniques.

1. Suppose an accident prone area is under surveillance and real time CCTV visuals are
available to you. Design a solution to automatically detect accidents on the road from those real time
CCTV visuals. Explain about any one pattern recognition algorithm you will make use here and how?

QP CODE:
Reg No: _______________
Name: _________________ PAGES: 2
Course Name: Pattern Recognition

PART A
1. In a town it was estimated that 3% of people have a particular disease. A diagnosis (5)
test was conducted for all the people, which yielded 8% false positive and 92%
true positive results. A person is found as positive after the test. What is the
probability that this person is truly having the disease?
2. How does morphological operations play a role in pattern recognition? (5)
3. How can visual imagery be analysed using convolutional neural networks? (5)
4. How does a decision tree handle continuous attributes? (5)
5. Define the terms: weights, bias, activations with respect to neural networks (5)
Part B
6. Illustrate the design principles of pattern recognition system with an example. (7)
7. Derive the fuzzy C spherical shells (FCSS) algorithm for the case that spherical (7)
clusters are to be identified.
8. (7)
9. (7)
10. Discuss the significance of pre-processing in feature selection. Illustrate any two (7)
methods used for pre-processing.
11. How can artificial neural networks be applied in Pattern recognition? Also (7)
illustrate the features of recurrent neural networks.
12. Construct a decision tree using the following data. (7)
Outlook Temp Humidity Windy Play Golf

Rainy Hot High False No
Rainy Hot High True No
Cloudy Hot High False Yes
Sunny Mild High False Yes
Sunny Cool Normal False Yes
Sunny Cool Normal True No
Cloudy Cool Normal True Yes
Rainy Mild High False No
Rainy Mild Normal False Yes
Sunny Mild Normal False Yes
Rainy Cool Normal True Yes
Cloudy Mild High True Yes
Cloudy Mild Normal False Yes
Sunny Hot High True No
Syllabus
Module 1: Introduction to Pattern Recognition
Basics of pattern recognition systems, various applications, Machine Perception,

classification of pattern recognition systems. Design of Pattern recognition system, Pattern
recognition Life Cycle. Statistical Pattern Recognition: Review of probability theory,
Gaussian distribution. Normal density and discriminant functions.
Module 2: Feature Selection
Feature selection – Outlier removal – Data normalization – Missing data, The Peaking
phenomenon, Feature selection using statistical hypothesis testing- Hypothesis testing basics
– Application of t-Test in feature selection. Class separability measures-Divergence-
Chernoff bound and Bhattacharya Distance-Scatter matrices, Feature subset selection –
Scalar feature selection, Feature vector selection.
Module 3: Clustering Algorithms
Unsupervised learning and clustering - Criterion functions for clustering. Cluster validation.
Fuzzy clustering algorithms- Point representatives- quadratic surfaces and representatives –
hyper plane representatives. Binary morphology clustering algorithms (BMCAs) –
Discretization – Morphological operations - Determination of clusters in a discrete binary
set- Assignment of feature vectors to clusters – The algorithmic scheme, Boundary detection
algorithms.
Module 4: Dimensionality reduction
Dimensionality reduction: Principal component analysis - its relationship to Eigen analysis.

Fisher discriminant analysis - Generalised Eigen analysis. Eigen vectors/Singular vectors as
dictionaries. Factor Analysis, Total variability space - a dictionary learning method.Non
negative matrix factorisation - a dictionary learning method.
Linear discriminant functions: Gradient descent procedures, Perceptron.
Module 5: Artificial neural networks and Pattern Classification
Artificial neural networks: Review of Artificial neural network concepts, convolutional

neural networks, recurrent neural networks.
Non-metric methods for pattern classification: non-numeric data or nominal data. Decision
trees: Classification and Regression Trees (CART).
Course Plan
No. Topic No. of
Lectures
(40
Hours)
1 Module 1: Introduction to Pattern Recognition 7
1.1 Basics of pattern recognition systems, applications 1
1.2 Machine Perception, Classification of pattern recognition systems 1
1.3 Design of Pattern recognition system 1
1.4 Pattern recognition Life Cycle 1
1.5 Statistical Pattern Recognition 1
1.6 Review of probability theory 1
1.7 Normal density and discriminant functions 1
2 Module 2: Feature Selection 10
2.1 Feature selection – Outlier removal 1
2.2 Data normalization – Missing data 1
2.3 The peaking phenomenon 1
2.4 Feature selection using statistical hypothesis testing 1
2.5 Hypothesis testing basics – Application of tTest in feature selection 1
2.6 Class separability measures-Divergence 1
2.7 Chernoff bound and Bhattacharya distance 1
2.8 Scatter matrices 1
2.9 Feature subset selection –Scalar feature selection 1
2.10 Feature vector selection 1
3 Module 3: Clustering Algorithms 9
3.1 Unsupervised learning and clustering 1
3.2 Criterion functions for clustering. Cluster validation. 1
3.3 Fuzzy clustering algorithms- Point representatives 1
3.4 Quadratic surfaces and representatives – hyper plane representatives. 1
3.5 Binary morphology clustering algorithms (BMCAs) 1
3.6 Discretization 1
3.7 Morphological operations - Determination of clusters in a discrete binary set 1
3.8 Assignment of feature vectors to clusters 1
3.9 The algorithmic scheme, Boundary detection algorithms. 1
4 Module 4: Dimensionality reduction 8
4.1 Principal component analysis - its relationship to Eigen analysis 1
4.2 Fisher discriminant analysis 1
4.3 Generalised Eigen analysis 1
4.4 Eigen vectors/Singular vectors as dictionaries 1
4.5 Total variability space - a dictionary learning method 1
4.6 Non negative matrix factorisation - a dictionary learning method 1
4.7 Linear discriminant functions: Gradient descent procedures 1
4.8 Perceptron 1
5 Module 5: Artificial neural networks and Pattern Classification 6
5.1 Review of Artificial neural networks, Introduction to deep neural networks 1
5.2 Convolutional neural networks 1
5.3 Recurrent neural networks 1
5.4 Non-metric methods for pattern classification: Non-numeric data or nominal 1
data
5.5 Decision trees: Classification and Regression Trees (CART) lecture 1 1
5.6 Decision trees: Classification and Regression Trees (CART) lecture 2 1
References
1. S.Theodoridis and K.Koutroumbas, “Pattern Recognition”, 4th Ed., Academic Press,
2009
2. C.M.Bishop, “Pattern Recognition and Machine Learning”, Springer, 2006
3. R.O.Duda, P.E.Hart and D.G.Stork, “Pattern Classification”, John Wiley, 2001
4. Hastie, T., Tibshirani, R. and Friedman, J. “The Elements of Statistical Learning”.

Springer. 2001.
221ECS011 BIOINFORMATICS CATEGORY L T P CREDIT
PROGRAM 3 0 0 3
ELECTIVE 2
Preamble:
This course helps the learners to provide practical/research solutions to problems in the
domain of Bioinformatics. It enables the learners to understand. concepts of Bioinformatics,
Application of AI in Bioinformtics, Big Data Bioinformatics and Data Analytics with NGS
data. This course helps the learners to develop practical solutions to problems in
bioinformatics.
Course Outcomes: The COs shown are only indicative. For each course, there can be 4 to 6
COs.
CO 1 Analyse basic concepts of Biomolecules, Biological databases, Sequence

characteristics (Cognitive Knowledge Level :Apply)
CO 2 Apply ML/DL Model for RNA /Protein structure Analysis (Cognitive Knowledge
Level :Apply)
CO 3 Apply Big data techniques in Bioinformatics (Cognitive Knowledge Level :Apply)
CO 4 Comprehend the Data Analytics pipelines for NGS data (Cognitive Knowledge
Level :Apply)
CO 5 Design and Develop RNASeq /Chip/ Metagenomics seq Pipelines (Cognitive
Knowledge Level :Apply)
CO 6 Develop/suggest a solution for any research problems in the field of Bioinformatics
(Cognitive Knowledge Level :Apply)

course.

program
CO 1
CO 2
CO 3
CO 4
CO 5
CO 6
Assessment Pattern
Apply 50-80%
Analyse 20-40%
Evaluate
Create
Mark distribution
Marks
100 40 60 2.5 hours



7 marks

1. Comprehend the nature and scope of Bioinformatics
2. Articulate the different biomolecules and the various databases
3. Comprehend the concept of sequence Alignments

1 Comprehend and apply ML/DL models for Bioinformatics data
2. Apply LSTM /GAN models
3. Comprehend Transformer based architecture

1. Comprehend the Big data concept, challenges and the various techniques in it.
2. Apply and Analyse Map Reduce concept with Hadoop and Spark
3. Apply Spark SQL for scaling genomic data

1. Comprehend the understanding of NGS pipelines
2. Articulate on Denovo assemblies

1. Demonstrate Understanding of RNASeq /Chipseq Pipelines
2. Apply the principle of QIIME

QP CODE:
Reg No: _______________
Name: _________________ PAGES : 4
Course Name: BIOINFORMATICS
PART A
1.
(i) Study the cartoon and comment on the heritable trait that
transmitted from parent to offspring
(ii) A structural gene has two DNA strands and for the partial sequence
of the gene given below, which is the template strand and why?
5’ACTCGATGCTAG3’
3’TGAGCTACGATC5’
2. Bioinformatics is the application of computational technology to handle the

rapidly growing repository of information related to molecular biology”. Give a
brief insight based on your knowledge on this definition of bioinformatics
3. Write short note on the need and application of Biological databases in sequence
analysis
4. Write a Python program using Pyspark to perform the following task.
• Create a dataframe with the following attributes student id, name, Marks
• Find the scholars whose mark <50% using spark sql comment
• Find the average and variance of the marks
• Find the maximum and minimum marks
5. Discuss the steps involved in the Metagenomics data analysis pipeline.
(5x5=25)
Part B
6. (a) 3 marks
For the given RNA molecule secondary structure, identify the (i) Stem (ii) Loop and
explain the importance of stem-loop structure of RNA.
(b) 4 marks
The protein synthesising process the shown below. Identify the different stages and
explain the role of mRNA?
7. (a) How do you submit biological data to a public database? List the major 3 marks
submission tools in NCBI.
(b) From NCBI, while downloading a gene sequence, you can download the 4 marks
sequence in two formats- GenBank & FASTA. To know more information
about the sequence, which format will be useful? Justify your answer.
8. (a) Align AGCTCAG with AGGTCA by using a suitable scoring scheme 5 marks
(b) 2 marks
Differentiate between Pairwise and Multiple sequence alignment.
9. (a) Write a simple python program for illustrating the map reduce model used in 4 marks
Hadoop?
Explain with examples the PageRank algorithm using Map Reduce 3 marks
programming concept
10. (a) Study the pictorial representation given below. Comment on the unknown 3 marks
block and its relevance in machine learning
(b) Explain architecture of CNN. Estimate the number of parameters using 2 4 marks
convolution layer model
11. (a) Briefly explain the big data challenges in the field of Bioinformatics. 3 marks
(b) Distinguish between Bowtie and BWA. 4 marks
12. (a) What is a contig? What is a scaffold? Describe de novo assembly. Name a 3 marks
software used for de novo assembly.
(b) Briefly explain the public sequence databases which support Next generation 4 marks
sequencing data and the various data formats being supported
Syllabus
Module 1: Introduction to Bioinformatics
Informational view of life science; Introduction to Biomolecules through games-Foldit

(Protein overlapping), Eterna Game (RNA Structure), History of Bioinformatics, Biological
databases, Sequence Similarity, identity and homology, Scoring Matices, Sequence
Alignment, Phylogeny
Module 2: AI in Bioinformatics
AI applications in the field of genomics, the role of deep learning and data mining in
computational biology and bioinformatics, ML/DL algorithms: ANN, CNN, LSTM, BERT,
GAN, Protein structure prediction (deep neural networks), RNA structure prediction (deep
learning models), RNA-protein binding sites prediction with CNN, Deep neural net to
predict target gene expression, Transcription Factor Binding via MLP, LSTM, CNN, Protein
Contextual Embeddings via BERT
Module 3 Big Data Bioinformatics
Overview of Big data , Data Storage and Analysis, Processing-SCV principle, Batch Vs
Stream processing, Big data Analytics- Typical Analytical Architecture -Types (Descriptive,
Inquisitive, Predictive, Perspective); Visualisation and Applications, Computational facilities
for analysing Big data – Cluster computing vs. Cloud computing, – Challenges in Big Data
Analytics , Big data Frameworks - MapReduce, Hadoop and Spark, Spark SQL and
dataframes, Spark for Bioinformatics, Big data analytics using Python- PySpark, Big data
processing for DNA sequence analysis – PASTASpark
Module 4 Data Analytics with NGS data
Introduction to next generation sequencing: NGS Platforms, NGS, advantages, limitations

and applications, NGS Data sources: NCBI SRA, EBI-ENA, DDBJ-SRA; SRA toolkit; NGS
Data analysis: FASTQ files, Quality check, Pre-processing, Mapping - Principles, tools -
BWA, Bowtie, SAM tools -output file formats -BAM, SAM; Denovo assemblies - Principles,
tools - SOAPdenovo, Velvet; Visualization tools - IGV; Whole Genome/Exome pipeline for
Variant calling - VCF files.
Module 5 Advanced Data Analytics with NGS data
RNAseq - Gene expression analysis, Differential expression analysis. Alternative splicing -

TopHat and Cufflinks for RNAseq, ChIPseq - Introduction and biological theories on
ChIPseq analysis. DNA fragment evaluation. Peak identification. Two condition comparison.
Saturation analysis. Motif finding and related theories, Metagenomics analysis using QIIME
and Picrust

(40 hrs )
1 Module 1 : Introduction to Bioinformatics
1.1 Informational view of life science; Definition; DNA-RNA and Protein as 1
information, Primary and secondary structure of DNA,
1.2 Chargaff’s Rules, Different forms of DNA, RNA, Introduction to 1
Biomolecules through games-Foldit (Protein overlapping), Eterna Game
(RNA Structure)
1.3 History of Bioinformatics, Definition of Bioinformatics, Bioinformatics 1
versus Computational Biology, Goals of Bioinformatics analysis,
1.4 Biological data bases :- File format, conversion of file format, Data 1
retrieval system, Genome browsers. Biological data file formats
1.5 Basic concepts of sequence similarity, identity and homology, Scoring
matrices- PAM and BLOSUM matrices,
1.6 Data retrieval system, Sequence databases-EMBL, GenBank, DDBJ;
1.7 Protein databases- UniProt, Protein Data Bank.
1.8 Concept of sequence alignment- pairwise and multiple Pairwise- Local 1
and global, ,
1.9 Dot plot, BLAST, Multiple sequence alignment (MSA) – CLUSTAL
Omega
1.10 Phylogeny: Basic concepts of phylogeny, Phylogenetic tree construction 1
using MEGA.
2 Module 2 : AI in Bioinformatics
2.1 AI applications in the field of genomics, the role of deep learning and 1
data mining in computational biology and bioinformatics.
2.2 ML/DL algorithms: ANN, CNN, LSTM, 1
2.3 ML/DL algorithms: BERT, GAN 1
2.4 ML model for protein expression /Sequence classification, 1
2.5 Protein structure prediction (deep neural networks), RNA structure 1
prediction (deep learning models),
2.6 RNA-protein binding sites prediction with CNN, Deep neural net to 1
predict target gene expression
2.7 Transcription Factor Binding via MLP, LSTM, CNN 1
2.8 Protein Contextual Embeddings via BERT 1
3 Module 3 Big Data Bioinformatics
3.1 Overview of Big data – Definition, Characteristics, Sources, Types- 1
Structured, Unstructured & Semi-structured;
3.2 Data Storage and Analysis -NAS, DAS, NoSQL databases; Processing- 1
SCV principle, Batch Vs Stream processing
3.3 Big data Analytics- Typical Analytical Architecture – Requirement for 1
new analytical architecture -Types (Descriptive, Inquisitive, Predictive,
Perspective); Visualisation and Applications,
3.4 Computational facilities for analysing Big data – Cluster computing vs. 1
Cloud computing, – Challenges in Big Data Analytics –Need of big data
frameworks
3.5 Big data Frameworks - MapReduce, Hadoop and Spark, Spark SQL and 1
dataframes,
3.6 Spark for Bioinformatics, Big data analytics using Python- PySpark, 1
3.7 Big data processing for DNA sequence analysis – PASTASpark 1
4 Module 4 Data Analytics with NGS data

4.1 Introduction to next generation sequencing: NGS Platforms, 1
4.2 NGS technologies (WGS, ChIP-seq & RNA-seq), advantages, limitations 1
and applications.
4.3 NGS Data sources: NCBI SRA, EBI-ENA, DDBJ-SRA; 1
4.4 SRA toolkit; NGS Data analysis: FASTQ files, Quality check, Pre- 1
processing,
4.4 Mapping - Principles, tools - BWA, Bowtie, SAM tools -output file 1
formats -BAM, SAM;
4.5 Denovo assemblies - Principles, tools - 1
4.6 SOAPdenovo, Velvet; Visualization tools - IGV; 1
4.7 Whole Genome/Exome pipeline for Variant calling - VCF files. 1
5 Module 5 Advanced Data Analytics with NGS data
5.1 RNAseq - Gene expression analysis,. 1
5.2 Differential expression analysis. Alternative splicing - 1
5.3 TopHat and Cufflinks for RNAseq 1
5.4 ChIPseq - Introduction and biological theories on ChIPseq analysis. 1
5.5 DNA fragment evaluation. Peak identification. Two condition 1
comparison. Saturation analysis.
5.6 Motif finding and related theories 1
5.7 Metagenomics analysis using QIIME and Picrust 1
Reference Books
 Brown, T.A. 2002 Genome. John Wiley Press, US.
 Campbell, A.M. & Heyer, L.J. 2002 Discovering Genomics, Proteomics and
Bioinformatics. Benjamin/Cummings
 Stuart M. Brown. Next-Generation DNA Sequencing Informatics, Second Edition.
New York University School of Medicine (ISBN-13: 978-1621921236)
 Xinkun Wang. Next Generation Sequencing Data Analysis, CRC Press. (ISBN13:
9781482217889).
 Mueller J. P. & Massaron L. (2016),”Machine learning for dummies”,John Wiley & Sons.
 Russell S. J. & Norvig P. (2003), “Artificial Intelligence A Modern Approach”, Pearson
Education.
 Jordan M. I. & Mitchell T. M. (2015), “Machine learning: Trends, Perspectives and
Prospects. Science”, 349(6245), pp. 255-260. (Journal Article)
 Kamath, U., Liu, J., & Whitaker, J. (2019). Deep learning for NLP and speech
recognition (Vol. 84). Cham: Springer.
 Gulli, A., Kapoor, A., & Pal, S. (2019). Deep learning with TensorFlow 2 and Keras:
regression, ConvNets, GANs, RNNs, NLP, and more with TensorFlow 2 and the Keras API.
Packt Publishing Ltd.
CODE HEURISTIC METHODS CATEGORY L T P CREDI
221ECS051 T
PROGRAM 3 0 0 3
ELECTIVE 2
Preamble:
The course introduces students to the concept of metaheuristics and heuristics-based
optimization algorithms. The course covers the topics gradient-based optimization,
population-based optimization algorithms, parallel methods and multi-objective and
combinatorial optimization algorithms. The course will enable students to use these
algorithms for solving hard problems.
Course Outcomes:
CO 1 Use gradient-based optimization algorithms to provide better solutions to

computational problems. (Cognitive Knowledge Level: Apply)
CO 2 Use population-based optimization algorithms to solve computational
problems(Cognitive Knowledge Level: Apply)
CO 3 Depict individuals in the problem using various representations (Cognitive
CO 4 Use parallel and coevolution methods to arrive at better solutions to computational
problems. (Cognitive Knowledge Level: Apply)
CO 5 Solve computational problems using multi-objective and combinatorial optimization
methods (Cognitive Knowledge Level: Apply)
course.

bachelor program

CO 1
CO 2
CO 3
CO 4
CO 5
Assessment Pattern
Apply 50-80%
Analyse 20-40%
Evaluate Assess through Assignments/
Projects
Create Assess through Assignments/
Projects
Mark distribution
Total CIE ESE ESE
Marks Duration
100 40 60 2.5 hours

Preparing a review article based on peer reviewed original publications (minimum 10

Course based task/Seminar/Data collection and interpretation : 15 marks
Test paper (1 number) : 10 marks
students.

having 5 marks for each question. (Such questions shall be useful in the testing of knowledge,
achievement and maturity of the students in a course, through long answer questions relating
to theoretical/practical knowledge, derivations, problem solving and quantitative evaluation),
with minimum one question from each module of which student should answer any five.
Each question can carry 7 marks.

1. Illustrate how hill climbing algorithm can be used to solve traveling salesman problem.
2. Your good friend is the groundskeeper at the mansion of mean old Mr. Mathis. In the
backyard, there is a huge fountain with a complex network of pipes controlled by over
100valves. One weekend, Mr. Mathis announces he’s going on vacation, and when he returns
he wants the fountain to spray as high as it can… but the plans for the pipe network have
beenlost! Plus, since she only has the weekend, your friend can’t possibly try all of the valve
combinations to find the optimal setting. Which local optimization search method might she
want to use in real life (since the fountain can’t be simulated on a computer) to maximize
theheight of the fountain? You may assume that a valve is either on or off, and the water
height is easily measured. If you need to make any other assumptions, state them clearly..

1. Consider the problem of finding the shortest route through several cities, such that each
city is visited only once and in the end return to the starting city (the Travelling Salesman
problem). Suppose that in order to solve this problem we use a genetic algorithm, in which
genes represent links between pairs of cities. For example, a link between London and Paris
is represented by a single gene ‘LP’. Let also assume that the direction in which we travel is
not important, so that LP = PL.
a) How many genes will be used in a chromosome of each individual if the number of cities
is 10?
b) How many genes will be in the alphabet of the algorithm?
2. Discuss how to use PSO for permutation problems and use the traveling sales person as an
example to illustrate your answer.
1. The orthogonal rectangle packing problem requires that rectangles are packed
into a bin with the aim of minimising the height of the packing.
a)Suggest two representations for this problem when using a genetic algorithm.
b) Present the advantage(s) and disadvantage(s) of each type of representation.
1. Illustrate the use of PSO with parallelism to generate multiple tests from a set of
question banks satisfying predefined levels of difficulty.
.Course Outcome 5 (CO5):
1. The Quadratic Assignment Problem (QAP) can be described as the problem of

assigning a set of facilities to a set of locations with given distances between the locations
and given flows between the facilities. The goal then is to place the facilities on locations in
such a way that the sum of the product between flows and distances is minimal. Describe
how the Ant Colony Optimization algorithm can be used to provide a solution to the QAP
problem.
QP CODE:
Reg No: _______________
Name: _________________ PAGES : 4
FIRST SEMESTER M. TECH DEGREE EXAMINATION, MONTH & YEAR
Course Name: HEURISTIC METHODS
Max. Marks: 60 Duration: 2.5 Hours
PART A
1. If we were using simulated annealing to solve the Travelling Salesman

Problem, suggest a neighbourhood function.
2. Explain the difference between a genotypic representation and a phenotypic

representation. Give an example of each.
3. Using an example, show why it is important to have a mutation operator in a

genetic algorithm.
4. Explain Front Rank Assignment by Non-Dominated Sorting
5. Discuss the Greedy Randomized Adaptive Search Procedures

(5x5=25)
Part B
6. (a) Describe the idea behind the simulated annealing algorithm making (4)
reference to its origins as an optimisation methodology.
(b) (3)
7. (a) There are five stages when planning to solve a problem using genetic (4)
programming. What are they? Give a short description of each.
(b) (3)
8. In the Travelling Salesman Problem we could represent the problem as (7)

a sequence of cities, 1..n.
i) Using a genetic algorithm with 1-point crossover we could infeasible
solutions. Assuming we could not change the crossover operator, how
could we overcome this problem?
ii) If we were able to use a different crossover operator, which one
would you suggest? Give an example.
9. Explain the steps in computing fitness values to express Pareto Strength (7)
10. Explain how Master-slave fitness assessment is performed in parallel (7)

metaheuristics.
11. Explain island models in evolutionary optimization processes. (7)
12. With reference to Ant Colony Optimisation (ACO) describe these (7)
algorithmic elements.
• Evaporation
• Visibility
• Transition Probability
Syllabus
Module 1 (Introduction and Gradient-based Optimization)
Metaheuristics-Introduction-Algorithms-Notation; Gradient based Optimization-Single state
methods-Hill Climbing-Single-State Global Optimization Algorithms -Adjusting the
Modification Procedure: (1+1), (1+λ), and (1, λ); Simulated Annealing
Module 2 (Population Methods)
Population Methods-Evolution Strategies-The Genetic Algorithm-Crossover and Mutation -

More Recombination -Selection; Exploitative Variations-Elitism -The Steady-State Genetic
Algorithm -The Tree-Style Genetic Programming Pipeline -Hybrid Optimization Algorithms-
Scatter Search -Differential Evolution -Particle Swarm Optimization
Module 3 (Representation)
Representation-Vectors -Initialization and Bias - Mutation-Recombination-Heterogeneous

Vectors-Phenotype-Specific Mutation or Crossover; Direct Encoded Graphs -Initialization
Mutation -Recombination; Trees and Genetic Programming-Initialization -Recombination-
Mutation -Forests and Automatically Defined Functions-Strongly-Typed Genetic
Programming-cellular Encoding-Stack Languages; Lists -Initialization -Mutation -
Recombination- Rulesets -State-Action Rules-Production Rules -Initialization -Mutation -
Recombination
Module 4 (Parallel Methods and Coevolution)

Parallel Methods-Multiple Threads-Island Models-Master-Slave Fitness Assessment -
Spatially Embedded Models; Coevolution-Population Competitive Coevolution -1-population
Competitive Coevolution-2 population Competitive Coevolution-N-Population Cooperative
Coevolution .
Module 5 (Multiobjective and Combinatorial Optimization)
Multi-objective Optimization - Naive Methods , Non-Dominated Sorting , Pareto Strength;

Combinatorial Optimization -General-Purpose Optimization and Hard Constraints -Greedy
Randomized Adaptive Search Procedures- Ant Colony Optimization -The Ant System -The
Ant Colony System-Guided Local Search .

1 Module–1 (Introduction and Gradient-based Optimization)
1.1 Metaheuristics-Introduction 1
1.2 Algorithms-Notation 1
1.3 Gradient based Optimization-Single state methods 1
1.4 Hill Climbing 1
1.5 Single-State Global Optimization Algorithms 1
1.6 Adjusting the Modification Procedure: (1+1), (1+λ) 1
1.7 Adjusting the Modification Procedure: (1,λ), Simulated Annealing 1
2 Module–2 (Population Methods)
2.1 Population Methods-Evolution Strategies 1
2.2 The Genetic Algorithm-Crossover and Mutation, More 1
Recombination -Selection
2.3 Exploitative Variations – Elitism- The Steady-State Genetic 1
Algorithm
2.4 The Tree-Style Genetic Programming Pipeline 1
2.5 Hybrid Optimization Algorithms, Scatter Search -Differential 1
Evolution
2.6 Particle Swarm Optimization 1
3 Module–3 (Representation)
3.1 Representation-Vectors -Initialization and Bias 1
3.2 Mutation-Recombination-Heterogeneous Vectors 1
3.3 Phenotype-Specific Mutation or Crossover 1
3.4 Direct Encoded Graphs -Initialization Mutation -Recombination 1
3.5 Trees and Genetic Programming-Initialization -Recombination- 1
Mutation
3.6 Forests and Automatically Defined Functions-Strongly-Typed 1
Genetic Programming
3.7 cellular Encoding-Stack Languages 1
3.8 Lists -Initialization -Mutation -Recombination- Rulesets - 1
3.9 State-Action Rules-Production Rules -Initialization -Mutation - 1
Recombination
4 Module– 4 (Parallel Methods and Coevolution)
4.1 Parallel Methods-Multiple Threads 1

4.2 Island Models-Master-Slave Fitness Assessment 1
4.3 Spatially Embedded Models 1
4.4 Coevolution-Population Competitive Coevolution 1
4.5 1-population Competitive Coevolution 1
4.6 2-population Competitive Coevolution 1
4.7 N-Population Cooperative Coevolution 1
5 Module–5 (Multiobjective and Combinatorial Optimization)

5.1 Multiobjective Optimization - Naive Methods 1
5.2 Non-Dominated Sorting , Pareto Strength 1
5.3 Combinatorial Optimization -General-Purpose Optimization and 1
Hard Constraints
5.4 Greedy Randomized Adaptive Search Procedures 1
5.5 Ant Colony Optimization -The Ant System 1
5.6 The Ant Colony System-Guided Local Search 1
Reference Books
1. Sean Luke, Essentials of Metaheuristics (Second Edition) Paperback,
lulu.com, June 21, 2013
2. Rajesh Kumar Arora, Optimization Algorithms and Applications, CRC Press,
Taylor and Francis Group
221ECS052 GAME THEORY CATEGORY L T P CREDIT
PROGRAM 3 0 0 3
ELECTIVE 2
Preamble: This course introduces Game Theoretic concepts and various strategic
environments. It covers mathematical foundations of game, Strategic form games, mixed
strategies, matrix games and Nash equilibrium and its theoretical concepts. This course helps
the learners to suggest solutions for computational problems using game theoretic approaches.
Course Outcomes:
CO 1 Make use of the concepts of linear algebra and probability to solve computational
problems. (Cognitive Knowledge Level: Apply)
CO 2 Explain the concepts of Game Theory. (Cognitive Knowledge Level: Understand)
CO 3 Use the concepts of Strategic Form Games, Mixed strategies and Matrix Games in
computational solving problems.(Cognitive Knowledge Level: Apply)
CO 4 Use the Nash equilibrium concepts in solving game theoretic problems. (Cognitive
CO 5 Develop solutions to computational problems using techniques/algorithms in Game
Theory (Cognitive Knowledge Level: Apply)

course.

bachelor program
CO 1
CO 2
CO 3
CO 4
CO 5
Assessment Pattern
Apply 50-80%
Analyse 0-40%
Evaluate
Create
Mark distribution
Total CIE ESE ESE
Marks Duration
100 40 60 2.5 hours


students.

7 marks

1. Let λ1, λ2,…, λn be the eigen values of the matrix A ϵ Rnxn. Show that the eigen values of
the matrix In - A are 1 - λ1, 1 - λ2, …, 1 - λn.
2. Solve the following problem using Simplex method (Show the tableau).
Maximize:
Subjected to:
1. Define a game. What is the difference between a non-cooperative game and a cooperative
game?
2. What is intelligence and how is it different from rationality?
1. There are n players. Each player announces a number in the set 1,2, ..., m where m is a
fixed positive integer. A prize of One Rupee is split equally between all the people whose
number is closes to two thirds of the average number. Formulate this as a strategic form
game.
2. Find the maxmin values, minmax values, maxmin strategies, minmax strategies of the
following game:
N = {1, 2}; S1 = S2 = {A, B}
U1 is 0,1,1,1 for (A,A), (A,B), (B,A), and (B,B) resp. U2 is 1,1,1,0 for (A,A), (A,B), (B,A),
and (B,B) resp.
1. Compute a Nash equilibrium for the two person game with S1 = {0, 1} , S2 = {3, 4} U1(x,
y) = −U2(x, y) = |x − y| ∀ (x, y) ∈ {0, 1} × {3, 4}.
2. First Price Auction: Assume two bidders with valuations v1 and v2 for an object. Their
bids are in multiples of some unit (that is, discrete). The bidder with higher bid wins the
auction and pays the amount that he has bid. If both bid the same amount, one of them gets
the object with equal probability 12 . In this game, compute a pure strategy Nash equilibrium
of the game.
QP CODE:
Reg No: _______________
Name: _________________ PAGES : 4
Course Name: GAME THEORY
Max. Marks : 60 Duration: 2.5

Hours
PART A
1. State the axioms of probability.

2. What is intelligence and how is it different from rationality?
3. A strategic form game has 2 players having 3 strategies each. What is the
minimum number and maximum number of pure strategy Nash equilibria for
such a game?
4. Using first principles, find all MSNE of the matching pennies game.
5. Show that a matrix A will have a saddle point if and only if the maxmin value is
equal to the minmax value. (5x5=25)
Part B
6. (a) Define the following terms and an example for each. (7)
1. Vector space
2. Linear independence
3. Basis
(b) Let λ1, λ2,…, λn be the eigen values of the matrix A ϵ Rnxn. Show that the
eigen values of the matrix In - A are 1 - λ1, 1 - λ2, …, 1 - λn.
7. (a) Solve the following problem using Simplex method (Show the tableau). (7)
Maximize:
Subjected to:
(b) Let S be any finite set with n elements. Show that the set ∆(S), the set of all
probability distributions over S, is a convex set.
8. (a) Define a game. What is the difference between a non-cooperative game and (7)
a cooperative game?
(b) There are n players. Each player announces a number in the set 1,2, ..., m
where m is a fixed positive integer. A prize of One Rupee is split equally
between all the people whose number is closes to two thirds of the average
number. Formulate this as a strategic form game.
9. (a) Braess paradox game: For the version without the link AB, we showed that (7)
any strategy profile with 500 players playing strategy A and 500 players
playing strategy B is a PSNE. Are there any other PSNEs?
(b) First Price Auction: Assume two bidders with valuations v1 and v2 for an
object. Their bids are in multiples of some unit (that is, discrete). The bidder
with higher bid wins the auction and pays the amount that he has bid. If both
bid the same amount, one of them gets the object with equal probability 12 .
In this game, compute a pure strategy Nash equilibrium of the game.
10. (a) Compute a Nash equilibrium for the two person game with S1 = {0, 1} , S2 (7)
= {3, 4} U1(x, y) = −U2(x, y) = |x − y| ∀ (x, y) ∈ {0, 1} × {3, 4}.
(b) Show that any strictly dominant (mixed) strategy in a strategic form game
must be a pure strategy.
11. (a) Find all mixed strategy Nash equilibria, applying the NASC, of the (7)
following game:
N = {1, 2}; S1 = S2 = {A, B};
U1 is 4,0,1,1 for (A,A), (A,B), (B,A), and (B,B) resp. U2 is 1,4,5,1 for
(A,A), (A,B), (B,A), and (B,B) resp.
(b) Find the maxmin values, minmax values, maxmin strategies, minmax
strategies of the following game:
N = {1, 2}; S1 = S2 = {A, B}
U1 is 0,1,1,1 for (A,A), (A,B), (B,A), and (B,B) resp. U2 is 1,1,1,0 for
(A,A), (A,B), (B,A), and (B,B) resp.
12. (a) Show that the payoff of any player under any correlated equilibrium is at (7)
least the maxmin value of that player.
(b) Show given any mixed strategy profile that we can always find a correlated
strategy that produces the same expected payoff to each player as the mixed
strategy profile.
Syllabus
Module 1 (Mathematical Preliminaries)
Probability theory – random variable(continuous and discrete), axioms, cumulative

distribution function, probability density function, Bayes rule, joint distribution. Linear
Algebra – vector space, span, independence, basis, rank. Set theory – open set, closed set,
convex sets etc. Optimization – basic concepts, simplex algorithm.
Module 2 (Introduction to Game Theory)
Introduction to game theory – trends and applications of game theory. Introduction to

Strategic form games – preferences, utilities, rationality, intelligence, classification of games.
Introduction to Extensive form games – definition, transform to strategic form.
Module 3 (Strategic Form Games)
Strategic form games – matching pennies, various forms like RPS, BOS, co-ordination,
prisoner dilemma, company dilemma games etc. Dominant strategy equilibria – strong, weak
and very weak dominance and illustrations. Pure strategy Nash equilibria – illustrative
examples, games without pure strategy Nash equilibria(NE), interpretations of NE, existence
of multiple NE, maxmin and minmax values, equilibria in extensive form games.
Module 4 (Mixed strategies and Matrix Games)
Mixed strategies – mixed strategy NE, properties, necessary and sufficient conditions,
maxmin and minmax values, domination. Utility theory – axioms, Von Neumann-
Morgenstern theorem and its computation, risk attitude of players. Matrix games – pure
strategies in matrix games, saddle points and pure strategy NE, mixed strategies in matrix
games, minmax theorem.
Module 5 (Nash Theorem and its Computation)
Existence of Nash equilibrium – correspondence and fixed point theorem, NE as a fixed

point. Existence conditions. Nash theorem – Sperner’s lemma, Brouwer’s fixed point
theorem. Computation of NE – supports, general algorithm for NE with examples.
Course Plan
No Topic No. of
Lectures ( 40)
1 Module 1 (Mathematical Preliminaries)
1.1 Probability theory – random variable(continuous and discrete), axioms, 3
cumulative distribution function, probability density function, Bayes rule,
joint distribution.
1.2 Linear Algebra – vector space, span, independence, basis, rank. Set 3
theory – open set, closed set, convex sets etc.
1.3 Optimization – basic concepts, simplex algorithm. 2
2 Module 2 (Introduction)
2.1 Introduction to game theory – trends and applications of game theory. 2
2.2 Introduction to Strategic form games – preferences, utilities, rationality, 3
intelligence, classification of games.
2.3 Introduction to Extensive form games – definition, transform to strategic 3
form.
3 Module 3 (Strategic Form Games)
3.1 Strategic form games – matching pennies, various forms like RPS, BOS, 3
co-ordination, prisoner dilemma, company dilemma games etc.
3.2 Dominant strategy equilibria – strong, weak and very weak dominance 2
and illustrations.
3.3 Pure strategy Nash equilibria – illustrative examples, games without pure 3
strategy Nash equilibria(NE), interpretations of NE, existence of multiple
NE, maxmin and minmax values, equilibria in extensive form games.
4 Module 4 (Mixed strategies and Matrix Games)
4.1 Mixed strategies – mixed strategy NE, properties, necessary and 3
sufficient conditions, maxmin and minmax values, domination.
4.2 Utility theory – axioms, Von Neumann-Morgenstern theorem and its 2
computation, risk attitude of players
4.3 Matrix games – pure strategies in matrix games, saddle points and pure 3
strategy NE, mixed strategies in matrix games, minmax theorem.
5 Module 5 (Nash Theorem and its Computation)
5.1 Existence of Nash equilibrium – correspondence and fixed point theorem, 3
NE as a fixed point. Existence conditions.
5.2 Nash theorem – Sperner’s lemma, Brouwer’s fixed point theorem. 2
5.3 Computation of NE – supports, general algorithm for NE with examples. 3
Reference Books
1. Y. Narahari. Game Theory and Mechanism Design. IISc Press and the World Scientific.
2014.
2. Michael Maschler, Eilan Solan, and Schmuel Zamir. Game Theory. Cambridge University
Press, 2013.
3. Sheldon M. Ross. A First Course in Probability. Pearson, Eighth Edition, 2010.
4. Gilbert Strang. Introduction to Linear Algebra. Wellesley-Cambridge Publishers, Fourth

Edition, 2009.
5. Vasek Chvatal. Linear Programming. W.H. Freeman & Company, 1983.

221ECS053 SOFTWARE PROJECT CATEGORY L T P CREDIT
MANAGEMENT PROGRAM 3 0 0 3
ELECTIVE 2
Preamble: This course provides fundamental knowledge in the Software Project Management. It
covers Software Development process, Software Quality Assurance, Project Management concepts
and technology trends. This course enables the learners to apply state of the art industry practices in
software development.
Demonstrate Traditional and Agile Software Development approaches (Cognitive

CO 1 Knowledge Level: Apply)
Prepare Software Requirement Specification and Software Design for a given

CO 2 Problem. (Cognitive Knowledge Level: Apply)
Justify the significance of design patterns and licensing terms in software

CO 3 development, prepare testing, maintenance and DevOps strategies for a project.
Make use of software project management concepts while planning, estimation,
CO 4 scheduling, tracking and change management of a project, with a traditional/agile
framework. (Cognitive Knowledge Level: Apply)
Utilize SQA practices, Process Improvement techniques (Cognitive Knowledge Level:
CO 5
Apply)
Program Outcomes (PO) : Outcomes are the attributes that are to be demonstrated by a
graduate after completing the course.

program

CO 1
CO 2
CO 3
CO 4
CO 5
Assessment Pattern
Apply 50-80%
Analyse 20-40%
Evaluate Assess through
Assignments
/Projects
Mark distribution
Marks
100 40 60 2.5 hours



7 marks

1. What are the advantages of an Boehm’s spiral model over a waterfall model?
2. How the process differs in agile software development and traditional software
development? Illustrate with a socially relevant case study. (Assignment question)
1. What is the relevance of building a requirements model in software development?

2. What are the characteristics of a good design? How do we assess the quality of a software
design?
3. Prepare a use case diagram for a bus ticket reservation system.

1. How do design patterns help software architects communicate the design of a complex system
effectively?
2. Differentiate between unit testing and integration testing.
3. Justify the need for DevOps practices?
1. What are the activities involved in software project management? Illustrate with an
example?
2. How risk analysis and management helps in the software development process?
3. What is the importance of project scheduling?
1. Why Software Process improvement is important?

2. What are the elements of Software Quality Assurance?
3. How SQA group helps in achieving high quality software?
QP CODE:
Reg No: _______________
Name: _________________ PAGES : 2
Course Name: SOFTWARE PROJECT MANAGEMENT
PART A
1. What are the elements of a software process? How do process models differ from
one another?
2. Why understanding the requirements of a problem is considered as one of the most
difficult tasks that face a software engineer? What are the steps required to
establish the groundwork for an understanding of software requirements?
3. Describe the formal and informal review techniques.
4. Differentiate between Problem-based Estimation, Process-based Estimation.
5. What are the elements of Software Quality Assurance? (5x5=25)
Part B
6. What are the advantages of an incremental development model over a
(a)
waterfall model? (7)
(b) Explain Agile Development techniques.
7. What is the importance of Requirements Engineering? Describe the distinct
(a) (7)
tasks involved in requirements engineering.
8. (a) Explain Continuous Integration, Delivery, and Deployment CI/CD/CD) (7)
9. (a) Describe the Software Configuration Process. (7)

10. Why technical reviews are considered as a good software engineering
(a)
practice? (7)
(b) What is a critical path? Illustrate its significance in a project schedule.
11. (a) Describe the CMMI process improvement framework. (7)

12. (a) Differentiate between GPL and LGPL?
(7)
(b) Describe COCOMO cost estimation model.
Syllabus
Module 1: Introduction to Software Engineering (8 hours)
Introduction to Software Engineering – The software Process – Process Models - Software

engineering ethics. Software process models - The Waterfall Model, Incremental Process Models –
Evolutionary Process Models – Concurrent Models, Specialized Process Models – Component-based
Development, The Unified Process, Product and Process. Agile development - Agile Principles –
Agile Process Models - Extreme Programming – SCRUM.
Module 2: Requirement Analysis and Design (8 hours)
Understanding Requirements – Requirements Engineering – Eliciting Requirements - Developing

Use Cases – Building the Requirements model - Validating Requirements, Requirements Analysis.
Design concepts - Design within the context of software engineering – Design Process- Design
concepts - Design Model. Template of a Design Document as per “IEEE Std 1016-2009 IEEE
Standard for Information Technology – System Design –. Software Design Descriptions “. Object-
oriented design using the UML, Design patterns.
Module 3 : Implementation and Testing (9 hours)
Implementation issues, Open-source development - Open-source licensing - GPL, LGPL, BSD.

Review Techniques - Cost impact of Software Defects, Review matrix and their use - Informal
Review, Formal Technical Reviews. Software testing strategies - Unit Testing – Integration Testing -
Validation testing - System testing , Debugging, White box testing - Path testing - Control Structure
testing, Black box testing, Testing Documentation and Help facilities.
Overview of DevOps and Code Management – Code management, DevOps automation, Continuous
Integration, Delivery, and Deployment (CI/CD/CD). Software Evolution - Evolution processes,
Software maintenance.
Software Project Management (8 hours)
Software Project Management Concepts- Estimation for Software Projects – Decomposition

Techniques- Problem-based Estimation, Process-based Estimation, Empirical Estimation Models –
COCOMO model , Project Scheduling, Risk management – Software Risks, Risk Identification, Risk
Projection, The RMMM plan. Software Configuration Management (SCM) – The SCM Repository -
The SCM Process.
Software Quality Assurance and Software Process Improvement (7 hours)

Software Quality - Software Quality Dilemma - Achieving Software Quality. Software Quality
Assurance - Elements of Software Quality Assurance - SQA Tasks – Statistical Software Quality
Assurance - Software Reliability. Software Process Improvement (SPI) - SPI Process - The CMMI
process improvement framework. ISO 9001:2000 for Software.
Course Plan
No. of
No Topic
Lectures ( 40)
1 Module 1 (Introduction to Software Engineering)
Introduction to Software Engineering – The software Process – Process
1.1 1
Models - Software engineering ethics.
1.2 Software process models - The Waterfall Model, Incremental Process Models 1
1.3 Evolutionary Process Models – Concurrent Models 1
1.4 Specialized Process Models – Component-based Development 1
1.5 The Unified Process, Product and Process 1

1.6
Agile development - Agile Principles – Agile Process Models 1
1.7 Extreme Programming 1
1.8 SCRUM Framework 1
2 Module 2 (Requirement Analysis and Design)
2.1 Understanding Requirements – Requirements Engineering 1
2.2 Eliciting Requirements - Developing Use Cases 1

Building the Requirements model - Validating Requirements, Requirements
2.3 Analysis 1
2.4 Design within the context of software engineering – Design Process 1
2.5 Design concepts - Design Model 1

Template of a Design Document as per “IEEE Std 1016-2009 IEEE Standard
2.6 for Information Technology Systems Design Software Design Descriptions” 1
2.7 Object-oriented design using the UML 1
2.8 Design patterns 1
3 Module 3 (Implementation and Testing)

Implementation issues, Open-source development - Open-source licensing -
3.1 1
GPL, LGPL, BSD
Review Techniques - Cost impact of Software Defects - Review matrix and
3.2 1
their use - Informal Review, Formal Technical Reviews
Software testing strategies - Unit Testing – Integration Testing, Validation
3.3 2
testing, System testing, Debugging
3.4 White box testing, Path testing, Control Structure testing 1
3.5 Black box testing, Testing Documentation and Help facilities. 1

Overview of DevOps and Code Management – Code management, DevOps
3.6 2
automation, Continuous Integration, Delivery, and Deployment (CI/CD/CD).
3.7 Software Evolution - Evolution processes, Software maintenance. 1
4 Module 4 (Software Project Management)
4.1 Software Project Management Concepts- Estimation for Software Projects 1

Decomposition Techniques- Problem-based Estimation, Process-based
4.2 1
Estimation
4.3 Empirical Estimation Models – COCOMO model 1
4.4 Project Scheduling 1

Risk management – Software Risks, Risk Identification, Risk Projection, The
4.5 2
RMMM plan.
4.6 Software Configuration Management (SCM) – The SCM Repository 1
4.7 The SCM Process 1
5 Module 5 (Software Quality Assurance and Software Process Improvement )
5.1 Software Quality - Software Quality Dilemma - Achieving Software Quality 1

Software Quality Assurance - Elements of Software Quality Assurance -
5.2 2
SQA Tasks
5.3 Statistical Software Quality Assurance - Software Reliability 1
Software Process Improvement(SPI) - SPI Process - The CMMI process
5.4 improvement framework. 2
5.5 ISO 9001:2000 for Software 1
Reference Books
1. Roger S. Pressman, Software Engineering : A practitioner’s approach, McGraw Hill
publication, Eighth edition, 2014.
2. Ian Sommerville, Software Engineering, Pearson Education, Tenth edition, 2015.
3. Ian Sommerville, Engineering Software Products: An Introduction to Modern
Software Engineering, Pearson Education, First Edition, 2020.
4. IEEE Std 830-1998 - IEEE Recommended Practice for Software Requirements
Specifications
5. IEEE Std 1016-2009 IEEE Standard for Information Technology—Systems Design—
Software Design Descriptions
6. David J. Anderson, Agile Management for Software Engineering, Pearson, 2003
7. Walker Royce, Software Project Management : A unified framework, Pearson
Education, 1998
8. Steve. Denning, The age of agile, how smart companies are transforming the way
work gets done. New York, Amacom, 2018.
CATEGORY L T P Credit
221LCS003 AI AND ML LAB
LAB 1 0 0 2 1
Preamble:
This course enables the learners to get hands-on experience in most popular supervised
learning algorithms, unsupervised learning algorithms and fundamental algorithms in the
area of Artificial Intelligence. This helps the learners to understand the process of
knowledge inference from raw data through dataset pre-processing and analysis. This
course also covers the AI based Algorithms, logical reasoning agents and implementation of
these reasoning systems using either backward or forward inference mechanisms.
CO# Course Outcomes
Implement machine learning algorithms using packages and libraries in Python for
CO1
various applications.(Cognitive Knowledge Level: Apply)
Implement python programs for supervised learning methods through Neural

CO2 network, Regression, classification and unsupervised learning through clustering
algorithms (Cognitive Knowledge Level: Apply)
Demonstrate real-world problems as state space problems, optimization problems

or
CO3
constraint satisfaction problems.
Simulate a given problem scenario, analyze its performance and develop

CO4 programming solutions for the problem scenario.
Design and develop an expert system by using appropriate tools and techniques.
CO5
Outcomes are the attributes that are to be demonstrated by a graduate after completing
the course.

program
PO1 PO2 PO3 PO4 PO5 PO6 PO7
CO1
CO2
CO3
CO4
CO5

The laboratory courses will be having only Continuous Internal Evaluation and carries 100
marks.
Final assessment shall be done by two examiners; one examiner will be a senior faculty from
the same department.
Continuous Evaluation : 60 marks
Final internal assessment : 40 marks
Lab Report:
All the students attending the Lab should have a Fair Report. The report should
contains details of experiment such as Objective, Algorithm/Design, Description,
Implementation, Analysis, Results, and Outcome. The report should contain a print
out of the respective code with inputs addressing all the aspects of the algorithm
described and corresponding outputs. All the experiments noted in the fair report
should be verified by the faculty regularly. The fair report, properly certified by the
faculty, should be produced during the time of the final assessment.
Syllabus
*Mandatory
1. Familiarization of basic Python Libraries such as Sklearn, Numpy, Pandas and Matplotlib.*
2. Implement and demonstrate Single, Multi variable and Polynomial Regression for a given
set of training data stored in a .CSV file and evaluate the accuracy.*
3. Implement a Python program to perform logistic regression on a dataset.
4. Write a Python program to implement Naive Bayes classifier and calculate the accuracy,
precision, and recall for your data set.*
5. Write a Python program to demonstrate the working of the decision tree based ID3
algorithm. Use an appropriate data set for building the decision tree and apply this
knowledge to classify a new sample.*
6. Assuming a set of data that need to be classified, use a Support Vector Machine classifier
to perform this task and evaluate the accuracy.*
7. Implement K-Nearest Neighbor algorithm to classify any dataset.
8. Implement K-Means Clustering using any given dataset.*
9. Build an Artificial Neural Network using Backpropagation algorithm and test the same
with appropriate dataset.*
10. Installation and working on various AI tools viz. Python, R, GATE, NLTK, MATLAB etc.*
11. Implement basic search strategies for selected AI applications*.
12. Implement state space search algorithms*
13. Implement informed search algorithms*
14. Implement backtracking algorithms for CSP*
15. Implement local search algorithms for CSP*
16. Implement propositional logic inferences for AI tasks*
17. Implementation of Knowledge representation schemes*
18. Implement travelling salesman problem*
19. Implementation of Game playing (adversarial search)
20. Mini Project that implement a real world application using AI techniques (Group project
with a maximum of four students)
( Use socially relevant dataset as far as possible )
Practice Questions
1. Review of Python programming: Programs using matplotlib / plotly / bokeh / seaborn.
2. Write a program to construct a Bayesian network considering medical data. Use this
model to demonstrate the diagnosis of heart patients using the standard Heart Disease
Dataset.
3. Implementation of Random Forest Classification in Python.
4. Write a program to implement k-Nearest Neighbor algorithm to classify the iris data set.
Print both correct and wrong predictions.
5. Program to implement text classification using Support vector machine.
6. Implement Naïve Bayes theorem to classify spam and non-spam emails.
7. Apply K-Means clustering to evaluate Student’s performance. The results expected show
the profile of a student with criteria for excellent performance, standard performance, and
nderperformance.
8. Implementation of Depth-First Search (DFS).
9. Write a program to implement water jug problem.
10. Implement variants of hill-climbing and genetic algorithms.
11. Implement tic tac toe game for 0 and X.
12. Develop a program to construct a pruned game tree using Alpha-Beta pruning. Take the
sequence, [5, 3, 2, 4, 1, 3, 6, 2, 8, 7, 5, 1, 3, 4] of MINIMAX values for the nodes at the cutoff
depth of 4 plies. Assume that branching factor is 2, MIN makes the first move, and nodes
are generated from right to left.
13. Write a program to implement heuristic search procedure.
14. Write a program to implement Expert system.
15. Write a program to implement search problem of 3 x 3 puzzles.
Reference Books:
1. Aurelien Geron, “Hands–On Machine Learning with Scikit–Learn and TensorFlow”, O’Relily.
2. Dan W. Patterson, “Introduction to AI and ES”, Pearson Education, 2007
3. Kevin Night, Elaine Rich, and Nair B., “Artificial Intelligence”, McGraw Hill, 2008
4. Patrick H. Winston, "Artificial Intelligence", Third edition, Pearson Edition, 2006
5. Artificial Intelligence by Example: Develop machine intelligence from scratch using real
artificial intelligence use cases -by Dennis Rothman, 2018
6. Brachman, R. and Levesque, H. 2004. Knowledge Representation and Reasoning, Morgan
Kaufmann.

Mtech AI Syllabus

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Mtech AI Syllabus

Uploaded by

Copyright:

Available Formats

SEMESTER I

CO 1 Analyse the Machine Learning concepts, classifications of Machine Learning

Program Outcomes (PO)

PO1: An ability to independently carry out research/investigation and developmentwork in

Mapping of course outcomes with program outcomes

Bloom’s Category End Semester Examination

Total CIE ESE ESE

100 40 60 2.5 hours

Continuous Internal Evaluation Pattern:

Continuous Internal Evaluation : 40 marks

End Semester Examination Pattern:

Course Level Assessment Questions

Course Outcome 2 (CO2)

2. A Poisson distribution is used to model data that consists of non-negative integers.

3. A uni-variate Gaussian distribution is used to model data that consists of non-negative

Course Outcome 4 (CO4):

7. How do we learn the conditional probability tables(CPT) in Bayesian networks if

Course Outcome 5 (CO5):

Reg No: _______________

Name: _________________ PAGES : 4

APJ ABDUL KALAM TECHNOLOGICAL UNIVERSITY

FIRST SEMESTER M.TECH DEGREE EXAMINATION, MONTH & YEAR

Course Code: 221TCS009

Course Name: MACHINE LEARNING

Max. Marks: 60 Duration: 2.5 Hours

Answer All Questions. Each Question Carries 5 Marks

1. Explain the principle of the gradient descent algorithm.

Find the maximum likelihood estimate(MLE) for θ.

Sl.No. Actual Predicted

Module-1 (Parameter Estimation and Regression) 8 hours

Overview of machine learning: supervised, semi-supervised, unsupervised learning,

Overfitting, Regularization techniques - LASSO and RIDGE. Classification algorithms:

Module-3 (Unsupervised learning) 8 hours

Unsupervised learning: clustering, k-means, Hierarchical clustering, Principal component analysis,

Module-4 (Support Vector Machine and Graphical Models ) 7 hours

Module-5 (Evaluation Metrics and Sampling Methods) 8 hours

Classification Performance Evaluation Metrics: Accuracy, Precision, Precision, Recall,

4.2 Support vector machines: Max margin classification

2. Kevin P. Murphy. Machine Learning: A Probabilistic Perspective, MIT Press 2012.

5. Tom Mitchell, Machine Learning, McGraw-Hill, 1997.

Mapping of course outcomes with program outcomes

100 40 60 2.5 hours

Continuous Internal Evaluation Pattern:

Continuous Internal Evaluation : 40 marks

Micro project/Course based project : 20 marks

Course based task/Seminar/Quiz : 10 marks

Test paper, 1 no. : 10 marks

The project shall be done individually. Group projects not permitted.

Test paper shall include minimum 80% of the syllabus.

Course Level Assessment Questions

1. Are the following sets of vectors linearly independent?

6. Find the singular value decomposition (SVD) of the following matrix

Course Outcome 2 (CO2)

Course Outcome 3(CO3):

1. Let J and T be independent events, where P(J)=0.4 and P(T)=0.7.

ii. Find P(J∪T)

ii. Find P(R>2).

i. A answers the first question,

ii. B answers the first question.

Course Outcome 4 (CO4):

1. Find the extrema of f(x, y) = x subject to g(x, y) = x2 + 2y2 = 3.

2. Maximize the function f(x, y, z) = xy + yz + xz on the unit sphere g(x, y, z) = x2 + y2