SUK · Third Year B.Tech. Computer Science and Engineering (Data Science) · NEP 2020
Machine Learning
DS3104 · Elective I · Semester V · 3 credits · ISE 10 + mid-semester 30 + ESE 60
One of the Elective I options.
Unit 1: Introduction to Machine Learning
7 hoursDefinition, terminology and types of learning; problem categories, architecture, process and life cycle; performance measures, tools and frameworks; data visualisation.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- What machine learning is
- Types of learning
- Models of machine learning
- Cross-validation and model evaluation
Unit 2: Regression
7 hoursSimple regression: hypothesis, cost function and parameter learning by gradient descent, the learning rate, regression in matrix form; multivariate linear regression, feature scaling and polynomial regression.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- Simple and multiple linear regression
- Gradient descent variants
- Polynomial regression
- Normalisation and standardisation
Unit 3: Classification: Logistic Regression and Neural Networks
8 hoursLogistic regression: hypothesis, decision boundary, cost function and gradient descent; multiclass classification; overfitting, underfitting and regularised linear and logistic regression; the neuron model, neural networks for multiclass classification and back-propagation.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- Logistic regression
- Binary and multiclass classification
- Bias-variance trade-off
- Ridge and lasso regularisation
- Perceptron and multilayer networks
- Backpropagation in neural networks
Unit 4: Classification: Decision Trees and Naive Bayes
7 hoursDecision trees: construction, common problems, algorithms and random forests; conditional probability and the naive Bayes classifier; the k-nearest-neighbours classifier.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- Decision tree learning and overfitting
- Bagging and random forests
- Bayesian learning and naive Bayes
- K-nearest neighbours
Unit 5: Unsupervised Learning
7 hoursClustering: k-means and hierarchical clustering; association rule mining.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- K-means and choosing k
- Hierarchical clustering
- Association rules and Apriori
Unit 6: Introduction to Graphical Models
6 hoursBayesian networks and Markov random fields; hidden Markov models and their training with the Viterbi and Baum-Welch algorithms.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- Bayesian networks
- Hidden Markov models
Unit contents are summarised from the official SUK syllabus . AiCanCode.org is not affiliated with Shivaji University, Kolhapur.