(SEM V) THEORY EXAMINATION 2020-21 MACHINE LEARNING TECHNIQUES

SECTION A (Any 2 Questions Explained Briefly)
 

Explain the concept of Machine Learning.
Machine Learning is a branch of artificial intelligence that enables computers to learn from data and improve their performance without being explicitly programmed. It focuses on developing algorithms that can identify patterns, make predictions, and take decisions based on experience.
 

What is the difference between linear and logistic regression?
Linear regression is used to predict continuous output values such as price or temperature, while logistic regression is used for classification problems where the output is categorical, such as yes/no or true/false. Logistic regression uses a sigmoid function to map values between 0 and 1.
 

SECTION B (Any 2 Questions Explained)
 

Explain K-Nearest Neighbour (KNN) algorithm.
KNN is a supervised learning algorithm used for classification and regression. It works by finding the K nearest data points to a given test instance and assigning the most common class among them. KNN is simple to implement and does not require a training phase, but it is computationally expensive for large datasets.
 

Explain the role of Genetic Algorithm in Machine Learning.
Genetic Algorithm is an optimization technique inspired by the process of natural selection. It works through selection, crossover, and mutation to find optimal or near-optimal solutions. In machine learning, genetic algorithms are used for feature selection, parameter optimization, and solving complex search problems.
 

SECTION C (Any 2 Questions Explained)
 

Why is SVM considered a large margin classifier?
Support Vector Machine is called a large margin classifier because it constructs a decision boundary that maximizes the distance between the separating hyperplane and the nearest data points from both classes. This margin maximization improves generalization and reduces classification error.
 

Explain the Confusion Matrix in Machine Learning.
A confusion matrix is a performance evaluation tool for classification models. It shows the number of correct and incorrect predictions by comparing actual and predicted values. It helps calculate accuracy, precision, recall, and F1-score, which are important metrics for evaluating classifiers.
 

Most Questions in This PDF Are Related To
 

Most questions in the Machine Learning Techniques (KCS-055) paper are related to supervised and unsupervised learning, regression and classification algorithms, decision trees, SVM, neural networks, reinforcement learning, genetic algorithms, Bayesian learning, confusion matrix, evaluation metrics, and optimization techniques. The paper mainly focuses on core machine learning algorithms and their theoretical understanding with practical relevance.