EI25402 Applied Machine Learning – Semester IV – EIE – R-2025

Subject Code & Name: EI25402 – Applied Machine Learning

Regulation: R-2025

Semester: IV (Fourth Semester)

Branch: B.E. Electronics and Instrumentation Engineering (EIE)

Credits / L-T-P: 3 Credits | L-T-P: 3-0-0

Course Objectives

• This course introduces foundational concepts and methods in machine learning, emphasizing supervised and unsupervised learning, regression, and classification.
• It equips students with practical skills in data preprocessing, algorithm development, and neural networks.
• The course also provides exposure to handling real-world datasets and understanding clustering and dimensionality reduction techniques.

Full Unit-wise Syllabus

Unit I – Introduction to Machine Learning

Objectives of machine learning, human learning vs machine learning, types of machine learning: supervised learning, unsupervised learning, regression, classification. The machine learning process: data collection and preparation, feature selection, algorithm choice, parameter and model selection, training, evaluation, bias-variance tradeoff, underfitting and overfitting problems.

Activity Based Learning: Explore machine learning applications in real-world scenarios, compare human learning vs machine learning through examples, simulate bias-variance tradeoff using sample datasets.

Unit II – Data Preprocessing

Data quality, data preprocessing: data cleaning, handling missing data and noisy data, data integration, redundancy and correlation analysis, continuous and categorical variables. Data reduction: dimensionality reduction using Linear Discriminant Analysis and Principal Components Analysis.

Activity Based Learning: Perform data cleaning on raw datasets, handle missing and noisy values, apply dimensionality reduction techniques such as PCA and LDA, analyze correlation between features in real datasets.

Unit III – Supervised Learning and Unsupervised Learning

Linearly separable and non-linearly separable populations, logistic regression, support vector machines: kernels, risk and loss functions, support vector machine algorithm, multi-class classification, support vector regression. Introduction to clustering, partitioning methods: K-means algorithm, mean shift clustering, hierarchical clustering, clustering using Gaussian Mixture Models. Clustering high- dimensional data: problems and challenges.

Activity Based Learning: Build and evaluate logistic regression models, implement and visualize support vector machines with different kernels, develop classification and regression models using SVM. Implement K-means and mean shift clustering on real-world datasets, perform hierarchical clustering and visualize dendrograms, cluster using Gaussian Mixture Models, explore challenges in high-dimensional clustering.

Unit IV – Neural Networks

Multi-layer perceptron, backpropagation learning algorithm, neural network fundamentals, activation functions, types of loss function, radial basis function network. Optimization: gradient descent algorithm, stochastic gradient descent, one case study.

Activity Based Learning: Develop a neural network model using a multi-layer perceptron, implement backpropagation and activation functions, apply optimization techniques like SGD, build and evaluate a neural network-based controller for a selected application.

Tasks

• T1: Develop a machine learning model using a public dataset available for an application of your choice.
• T2: Carry out data preprocessing tasks such as handling missing/outliers, dimensionality reduction, etc., on the public raw datasets available.
• T3: Perform the correlation analysis between the various inputs and outputs of a system.
• T4: Carry out the clustering on real-world datasets using any two popular clustering algorithms and comment on their performance.
• T5: Build a multilayer neural network-based controller for a simulated nonlinear process of your choice.

Course Outcomes (COs)

• CO1: Explain the fundamental concepts and processes of machine learning, including supervised and unsupervised learning, regression, classification, and bias-variance tradeoff.
• CO2: Preprocess and analyze datasets, handling missing/noisy data, feature selection, and dimensionality reduction using techniques like PCA and LDA.
• CO3: Design, implement, and evaluate supervised and unsupervised learning models, including logistic regression, SVMs, and clustering algorithms such as K-means and hierarchical clustering.
• CO4: Develop and train neural network models using multi-layer perceptrons, backpropagation, and optimization techniques for classification and regression applications.

Assessment Pattern (Quick Note)

• Weightage: Continuous Assessment 40% | End Semester Examinations 60%
• Internal methodology: Assignments (20%), Solution to application-oriented problems using software (20%), Solving of GATE questions (20%), Internal Examinations (40%)

Source: Official Anna University – B.E. Electronics and Instrumentation Engineering R-2025 Syllabus
Last Updated: October 2026