Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

WEEK 4

TEAM: CODE-BREAKERS
Problem Statement

Naïve Bayesian Classifier(Dataset)

Problem Description
Given a dataset where each instance is characterized by a set of features and a corresponding class label,
the task is to build a Naïve Bayesian Classifier to predict the class label of new instances based on their
features. This involves training the model on a portion of the dataset to estimate the probabilities of
different classes based on feature values and then evaluating its accuracy on the remaining data to ensure
it can make reliable predictions.

Project Outcome

Scan for Code Medium Video Product


WEEK 5
TEAM: CODE-BREAKERS
Problem Statement

Naïve Bayesian Classifier Model

Problem Description
Given a dataset with instances characterized by a set of features and corresponding class labels,
the task is to build a Naïve Bayesian Classifier. This classifier should predict the class label of
new instances based on their features. The model will be trained on a portion of the dataset to
estimate the probabilities of different classes based on the features, and then its accuracy will be
evaluated on the remaining data to ensure reliable predictions.
Project Outcome

Scan for Code Medium Video Product


WEEK 6
TEAM: CODE-BREAKERS
Problem Statement

Bayesian Network

Problem Description
Given a set of variables with probabilistic relationships among them, the task is to construct a
Bayesian Network to model these dependencies. The Bayesian Network should be able to represent
the conditional dependencies via a directed acyclic graph (DAG) and include conditional probability
tables (CPTs) for each variable. The objective is to use this network to perform probabilistic inference,
answering queries about the likelihood of certain events given observed evidence. This involves
identifying the variables, structuring the network, defining the CPTs, and utilizing the network for
inference in practical applications such as diagnostic systems, decision support, and risk assessment.

Project Outcome

Scan for Code Medium Video Product


WEEK 7
TEAM: CODE-BREAKERS
Problem Statement

EM ALGORITHM (K MEANS ALGORITHM)

Problem Description

The K-Means algorithm, a specific application of the Expectation-Maximization (EM)


algorithm, aims to partition a dataset into \( K \) clusters by iteratively assigning points to the
nearest cluster center (Expectation step) and updating cluster centers to minimize intra-cluster
variance (Maximization step). It repeats these steps until convergence, ensuring each point
belongs to the cluster with the closest mean, optimizing clustering quality based on distance
metrics.

Project Outcome

Scan for Code Medium Video Product


WEEK 8
TEAM: CODE-BREAKERS
Problem Statement

K- NEAREST NEIGHBOUR ALGORITHM

Problem Description
The K-Nearest Neighbor (K-NN) algorithm is used for classification and regression tasks
based on labeled training data. For classification, given a new data point, K-NN identifies the
K nearest neighbors in the training set and predicts the class label based on the majority class
among these neighbors. In regression, it predicts the output value for the new data point based
on the average of the values of its nearest neighbors. The choice of \( K \) (number of
neighbors) and the distance metric (e.g., Euclidean, Manhattan) significantly influence the
algorithm's accuracy and computational efficiency.
Project Outcome

Scan for Code Medium Video Product


WEEK 9
TEAM: CODE-BREAKERS
Problem Statement
LOCALLY WEIGHTED REGRESSION ALGORITHM

Problem Description
Locally Weighted Regression (LWR) predicts \( y \) for new input \( x \) by emphasizing nearby data points
more than distant ones. It computes predictions using weighted least squares, where weights \( w^{(i)} \) are
determined by a kernel function based on the distance from \( x \). LWR adapts locally to data patterns but
requires tuning a bandwidth parameter \( \tau \) and can be computationally intensive for large datasets
due to distance computations for each prediction.

Project Outcome

Scan for Code Medium Video Product


Week 4

Week 5

Week 6
WEEK 7

WEEK 8

WEEK 9

You might also like