ML Sample PDF

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5

Page

1| www.notesera.in
multi-class common Some others. among faces,
categorizing
andspecies, classifying images, classifying example, Forlabel. one only
assigned
to outcome
is each Here outcomes. abnormal and normal
concept
of the have not does outcomes
and wothan more with
problemclassification of
type:Multi-class
ais Classification Multi-Class 2.
machines. vector support and bayes,
simple trees,decisionregression, logistic algorithms
are classification
binary common Some etc. spam, not spamor false, or
true no, oryes
example, outcomes.
For possible two only has thatlearning machine
classificationin problem
in of
type :Binary
aClassification
is Binary 1.
below.described areclassification,
which
imbalanced classification,
and multi-labelclassification, multi-class
classification, binary algorithms
ar:e classification common most The
Learning: Machine Classification
tasks
in Types
of
variables. more orone based
on classes into data input of
set given
categorizinga process
of learning machine supervised Classification
a is
inteligence.
artificial and gaming computer field
of the pioneer in employee and
IBM Samuel,
an Arthur by1959 coined
in learning
was machine term The
learning? machine father
of the Whiso
problems. solve humans how sinilar
to is
that way atasks
incomplex perform to
used systems
are intelligence Artificial
behavior. human intelligent imitate machine
to capability
aof the defined
as
broadly intelligence,
whichis artificial subfield
of leaming
ais Machine
learning. machine with
history rich ahas accuracy.
IBM improving
its gradually learn, humans that
way thimitate
e algorithms
to and data ofuse the focUses
on which science
computer andinteligence
(AI) artificial branch
of learning
ais Machine
learning? machine What
is
UNI1T
function Nearest-Neighbors,
Gaussian
Regression, Logistic descent; Gradiernt function, Regression:
Cost Linear
Leaning: SupervisedClassification; Applications, learning. machine Introduction to
Unit-1 Content:
Uni1t Learning Machine
Machine Learning Unit-2

Content : Unit-2
Overfitting And Underfitting, Regularization, Bias And Variance, Decision Trees, NarVe
Bayes, Support Vector Machines, Kernel Methods.

UNIT-2
What are Overfitting and Underfitting ?
Ans. Overfitting and Underfitting are the two main problems that occur in
machine learning and degrade the performance of the machine learning
models.
The main goalof each machine learning model is to generalize well. Here
generalization defines the ability of an ML model to provide a suitable output by
adapting the given set of unknown input.
Before understanding the overfitting and underfitting, let's understand some basic
term that will help to understand this topic wel:
Signal : In machine learning, the term "signal" refers to the useful
information or patterns in the data that the model is trying to learn.
Noise : Noise is unnecessary and irelevant data that reduces
the performance of the model.
Bias : It measures the difference between the model's prediction and the target value.
If the model is oversimplified, the predicted value would be far from the ground truth
resulting in more bias. A model with high bias is said to be underfitting, meaning it is too
simple and unable to capture the complexity and patterns in the data.
Variance : If the machine learning model performs well with the training
dataset, but does not perform well with the test dataset., then variance
OCCUrs. Machine Learning
Errors

Reducitble Error Irreducible Error

Bias

Variance

1. Reducible errors: These erors can be reduced to improve the


model accuracy. Such errors can further be classified into bias
and Variance.
2. Irreducible errors: These errors wll always be present in the model
regardless of which algorithm has been used. The caUse of these
errors is unknown variables whose value can't be reduced.
22
www.notesera.in
Machine Learning Unit - 3

Content: Unit-3
Unsupervised Learning: Clustering: K-means, Dimensionality Reduction: PCA, Matrix
Factorization and Matrix Completion, Ranking, Recommender System.

Unit -3
What is unsupervised learning ?
" Unsupervised learning is a type of machine learning that learns from
unlabeled data. This means that the data does not have any
pre-existing labels or categories.
" The goalof unsupervised learning is to discover patterns and
relationships in the data without any explicit guidance.

Unsupervised Machine Learning


Output

Apples
Input Raw Data Algorithm

Unlabeled Data
Interpretation Processing
Oranges

Let's take an example of Unsupervised Learning for a baby and her family
dog. CET OUT OE YOUR COMIORT ZONE

Baby has not seen this dog before. But she recognizes many features (2 ears,
eyes, walking on 4 legs) are like her pet dog. She identifies the new animal as
a dog. This is unsupervised learning, where you are not taught but you lean
from the data (in this case data about a dog.)
Had this been supervised learning, the family friend would have told the baby
that it's a dog as shown in the above Unsupervised Learning example.
Why Unsupervised Learning?
1. It finds allkinds of unknown patterns in the data.
2. Unsupervised methods help you to find features which can be useful for
categorization.
3. In real-time unsupervised learning, learners analyze data as it comes in
and labels it without any direct instructions.
Key Points of Unsupervised Machine Learning:
1. Clustering algorithms group similar data points together based on their
inherent characteristics.

www.notesera.in Page | 39
Machine Learning Unit - 4

Content: Unit-4
Introduction to Neural Network, Perceptron, Feed forward, Back Propagation,
Recurrent Neural Network. Introduction to Python machine learning libraries: Keras,
Tensorflow and Theano.

Unit -4
What is a Neural Network ?
A Neural Network in machine learning is a computational model inspired by
the structure and functioning of the human brain. It is composed of
interconnected nodes, or artificial neurons, organized into layers. Neural
networks are used for various tasks, including pattern recognition,
classification, regression, and other machine learning problems.
A neural network is a machine learning program, or model, that makes
decisions in a manner similar to the human brain, by using processes that
mimic the way biological neurons work together to identify phenomena.
Here are key components and concepts associated with neural networks:
1. Neurons (Nodes) : Neurons are the basic building blocks of a neural
network. Each neuron processes information and makes a decision
based on its input, In an artificial neural network, a neuron takes input
values, applies weights. performs a weighted sum, and passes the result
through an activation function.
2. Layers : Neural networks are organized into layers, which are groups of
neurons. The most common layers include:
" Input Layer: Receives input features.
" Hidden Layers: Intermediate layers between the input and output
layers, where computations are performed.
Output Layer: Produces the final output or prediction.
3. Weights and Bias: Each connection between neurons has an
associated weight, representing the strength of that connection.
Additionally, each neuron has a bias, which allows the network to
capture non-linear relationships in the data.
4. Activation Function : The activation function compares the input value
to a threshold value. If the input value is greater than the threshold
value, the neuron is activated. It's disabled if the input value is less than
the threshold value, which means its output isn't sent on to the next or
hidden layer.

www.notesera.in Page| 60
Machine Learning Unit 5

Content: Unit-5
Evaluating Machine Learning algorithms and Model Selection, Ensemble
Methods: Boosting. Bagging, Random Forests, Deep learning, Semi-supervised
Learning, Reinforcement Learning.
Unit- 5
Evaluating Machine Learning algorithms :
Evaluating machine learning algorithms means figuring out how good they are
at their job. It's like testing a student to see how well they've learneda subject.
We use different tests (metrics) to measure their performance, like accuracy or
error rate. We give the algorithm some data to leam from (training data) and
then see how well it does with new data (testing data). If it does well, it means
it's learned effectively and can be trusted to make predictions or decisions.
Define classification of metrics ?

Classification Metric is a number that measures the performance of your


machine learning model when it comes to assigning observations to certain
classes. Binary classification is a particular situation where you just have two
classes: posifive and negative. ceTouT OF YOUR COMFORT ZONE
Performance Metrics for Classification :

Evaluating the performance of a Machine learning model is one of the


important steps while building an effective ML model. To evaluate the
performance or quality of the model, different metrics are used, and these
metrics are known as performance metrics or evaluation metrics.
These performance metrics help US understand how well our model has
performed for the given data. In this way, we can improve the model's
performance by tuning the hyper-parameters. Each ML model aims to
generalize well on unseen/new data, and performance metrics help determine
how well the model generalizes on the new dataset.
SuPERVtSED LEARNENG

CLASSIFcATEOW RESRESsstON

ACCURAcY

PRECTStON RECALL

www.notesera.in Page | 76

You might also like