Course Title: Data Science & Engineering
Course Code: DS1010
L T P C: 1-0-0-1
Category: Core UG
Prerequisite Courses: None
Content: A walk through different real-world applications drawn from the fields of Computer Vision, Speech Processing, natural language processing, Bioinformatics, AI Systems etc., connecting it to the theory and engineering aspects of Data Science.
Course Title: Introduction to Optimization
Course Code: DS2010
L T P C: 3-0-0-3
Category: Core UG
Prerequisite Courses: Introduction to Programming, Mathematics I - Linear Algebra, Mathematics II - Multivariate Calculus
Content: Revision of Prerequisites, Linear Algebra and multivariate derivative, Characterization of maxima and minima: Conditions of maxima or minima for constrained and unconstrained Problems; Application of Linear Programming and simplex; Major Algorithms in unconstrained Optimization - Newton, quasi Newton, Steepest descent line search methods; Application of Constrained Application; Major Algorithms in constrained optimization with application [eg., barrier, Penalty, augmented lagrangian, interior point methods; Large Scale Optimization; Discrete Optimization [eg., combinatorial and integer programming]; Major Algorithms in derivative free-methods [eg., simulated annealing, bayesian optimization, surrogate-assisted optimization].
Course Title: Data Structures and Algorithms for Data Science
Course Code: DS2030
L T P C: 3-0-3-5
Category: Core UG
Prerequisite Courses: Introduction to Programming
Content: Introduction to Object Oriented Programming, Arrays, and Linked Lists; Analysis Tools, Recursion; Stacks and Queues; Tree Structures, Heaps, Sorting, Priority Queues, Divide and Conquer strategy; Hashing functions, Bloom Filters; Search Tree Structures; String processing, Dynamic programming; Graphs, greedy algorithms (Graph Traversal Algorithms).
Course Title: Introduction to Artificial Intelligence
Course Code: DS2020
L T P C: 3-0-2-4
Category: Core UG
Prerequisite Courses: Introduction to Programming, Discrete Mathematics
Content: Introduction to AI - Rational Agents; Search-(BFS, UCS, A* Search, Heuristics, Local Search); Adversarial Search; Knowledge Representation and logical inference (propositional logic, resolution, predicate logic, ontologies); Planning; Probabilistic reasoning (probabilistic graphical model, exact and approximate inference); Markov Decision Processes (Introduction to MDP); Decision making under uncertainty (Introduction to PoMDPs, game theory, mechanism design); Reinforcement learning (Introduction to RL).
Course Title: Computer Systems for Data Science
Course Code: DS2040
L T P C: 3-0-3-5
Category: Core UG
Prerequisite Courses: Data Structures and Algorithms for Data Science
Content: ORGANIZATION OF COMPUTING SYSTEMS: Introduction, Introduction and Data representation, processing unit, memory and I/O subsystems; SYSTEMS PROGRAMMING: Kernel, virtual memory, exceptions, processes, files, threads, scheduling, List potential threats to operating systems (eg., software vulnerabilities, authentication, issues, malware) and the types of security features designed to guard against them; PROGRAM EXECUTION: How programs are executed on a machine (with a particular focus on linux-based operating systems); Program segments/sections; The ELF Format; Linking and loading; Linux dynamic libraries (shared objects); Multitasking and paging; Address translation; Memory Protection; ADDITIONAL TOPICS: Threads, virtualization, cloud computing, CUDA, security vulnerabilities and hardening.
Course Title: Database Systems
Course Code: DS3020
L T P C: 3-0-3-5
Category: Core UG
Prerequisite Courses: Data Structures and Algorithms for Data Science
Content: Introduction: Database applications and purpose, Data abstraction and manipulation, Relational databases, Database schema, Keys, Relational query languages, algebra, tuple and domain calculus; SQL: Data definition, basic SQL query structure, set operations, nested subqueries, aggregation, null values, database modification, join expressions, views, Integrity constraints in SQL, triggers; Database Design: Entity-relationship model, reduction to relational schema, E-R design issues, Relational Database Design: Features of good design, Functional dependency theory, Decomposition using functional dependency, Normal forms (1NF, 2NF, 3NF, BCNF), Algorithms for decomposition; Storage Management and Indexing: Overview of secondary storage, storing tables, Index concept, clustered and nonclustered indices, B+-tree indices, hash indices, bitmap indices; Query processing and optimization: Evaluation of relational algebra expressions, query equivalence, join strategies, query optimization algorithms; Transactions: Concept, ACID properties, Transactions SQL statements; Concurrency control: Lock-based protocols, 2-phase locking, Deadlock handling, Multiple granularities, Timestamp-based protocols, Multi-version protocols, concurrency control for index structures; Recovery: Failure Classification, Recovery, and atomicity, Recovery algorithms, Buffer management; Introduction to modern databases: Different types of NoSQL databases, their advantages, and disadvantages.
Course Title: Introduction to Machine Learning
Course Code: DS3010
L T P C: 3-0-3-5
Category: Core UG
Prerequisite Courses: Introduction to Optimization, Probability and Statistics
Content: Introduction to the course, recap of linear algebra (vector derivative) and probability theory (Bayes Rule) basics; Regression, ridge regression; Classifier - Linear Classification (e.g., perception, maximum margin, logistic regression), Non linear classification (e.g., KNN, use of Kernel in SVM); Evaluation and Model Selection: ROC Curves, Evaluation Measures, Cross validation, Significance tests; Ensemble Methods: Boosting, Bagging, Decision Trees, Random Forests; Feature extraction (Principal Component Analysis, Canonical Correlation Analysis); Clustering: K-mean, Gaussian Mixture Model, Expectation Maximization, density based clustering; Sequential learning (HMM).
Course Title: Data Analytics
Course Code: DS3030
L T P C: 2-0-3-4
Category: Core UG
Prerequisite Courses: Data Structures and Algorithms for Data Science, Linear Algebra, Probability and Statistics, Artificial Intelligence
Content: Introduction: data definition, different types of data, data collection, data storage, data management, data driven decision making and system development; Descriptive and Inferential statistics: Measures of Central tendency, Measures of dispersions, analysis of variance, correlation analysis; Data Cleaning: Noise removal, outlier detection, missing value handling, feature engineering with cases of graph and texts, feature selection, normalization, standardization; Data transformation, dimensionality reduction (PCA, t-SNE, auto-encoder), regression; Association rule mining, pattern recognition using K-means clustering, hierarchical clustering, bi-clustering, density based clustering; Data visualization: table, graph, histogram, pie-chart, area-plot, box-plot, scatter-plot, bubble-plot, waffle charts, word clouds; Introduction to Experimental Design, Basic Analysis Techniques: Statistical hypothesis generation and testing, Chi-square test, Maximum likelihood test; Case study: IPL/Covid/Traffic.
Course Title: Introduction to Deep Learning
Course Code: DS3040
L T P C: 3-0-3-5
Category: Core UG
Prerequisite Courses: Introduction to Optimization, Machine Learning
Content: Revision of Linear Algebra, Probability, Optimization; Basics of learning: model, weight, data, loss functions, learning, prediction, output functions, connection with gradients and optimization, linear and non-linear classifiers; Perceptron: model, algorithm, basic neuron, XOR; Feed forward neural network [FNN] - Multi-layer perceptrons, weights, learning problem, Forward propagation, Backpropagation, Regularization methods, dropout, batchnorm, data augmentation; Convolutional neural network (CNN) - Convolution, filters, padding, pooling, CNN for image classification, Example architectures, various applications; Recurrent neural network (RNN) - Recurrence, various types of RNNs for different applications - Drawbacks of RNNs, Long Short Term Memory networks (LSTM), GRU, Attention and memory networks; Auto-encoders - Encoder-decoder, seq2seq case study, dimensionality reduction, denoising, Representation learning; Generative adversarial network (GAN), WGAN, conditional-GAN; Transformers.
Course Title: AI Ethics
Course Code: DS3060
L T P C: 2-0-0-2
Category: Core UG, Elective PG
Prerequisite Courses: Introduction to Artificial Intelligence, Machine Learning
Content: What is AI Ethics; Potential Harms Caused by AI Systems; Individual and social impact of ML and AI; Support, Underwrite, and Motivate (SUM Values): Respect, Connect, Care, and Protect; The FAST Track Principles - Fairness: data, design, outcome implementation, Accountability: answerability and auditability, Sustainability: Stakeholder Impact Assessment, accuracy, Reliability,security, and robustness, Privacy, Data poisoning, adversarial attack, Transparency: interpretable AI.
Course Title: Data Engineering
Course Code: DS5003
L T P C: 1-0-3-3
Category: Core PG
Prerequisite Courses: None
Content: Data Collection: Various sources and types of data: text, video, audio, biology etc; Data Preprocessing: Cleaning data, missing data imputation, noise elimination, feature selection and dimensionality reduction, normalization; Data Storage: Database, Schema, ER diagram, SQL, functions, stored procedures, indexing B+tree, MongoDB, Client-Server Architecture; Information Retrieval: index construction, scoring models, complete search engine mechanism, evaluation methods; Data Processing: Data structures. Stack, Queue, Linked List, Associated memory, Graphs. Algorithms. Searching, Sorting, Graph traversal, Complexity; Data Analysis: regression, principal component analysis, canonical correlation analysis, analysis of variance; Data Visualization: table, graph, histogram, pie-chart, area-plot, box-plot, scatter-plot, bubble-plot, waffle charts, word clouds.
Course Title: Optimization
Course Code: DS5005
L T P C: 3-0-0-3
Category: Elective UG/PG
Prerequisite Courses: None
Content: Introduction: Motivation and examples; Basics: Rd, vectors, matrices, norm, sequences & convergence, functions in one and several variables, Taylor series, derivatives, gradient, sub-gradient, Hessian, properties of symmetric operators, contours, affine functions, hyper-planes, convex functions, minima: local and global, subspaces, affine spaces, half-spaces, convex sets; Unconstrained Optimisation: gradient descent, line search, rates for various classes of convex functions, steepest descent, Newton’s method, conjugate gradient method, quasi-Newton method, linear least-squares regression: rates; Constrained Optimisation: linear and convex constraint sets, linear programming, simplex, interior-point method, duality theory: primal/dual programs, weak, strong duality, KKT conditions; Stochastic Optimisation: stochastic gradient descent, step-size conditions, Keifer-Wolwowitz method, simultaneous perturbation stochastic approximation (SPSA) method, smoothed functional method.
Course Title: Big Data Lab
Course Code: DS5102
L T P C: 1-0-3-3
Category: Core PG, Elective UG
Prerequisite Courses: None
Content: Lab on set up: manipulating files in HDFS; Basic programs of Hadoop MapReduce: Driver code, Mapper code, Reducer code, RecordReader, Combiner, Partitioner; Pig: Introduction to PIG, Execution Modes of Pig, Comparison of Pig with Databases, Grunt, Pig Latin, User Defined Functions, Data Processing operators; Big data analytics in Spark using PySpark: Installing Apache Spark, Spark Ecosystem, Resilient Distributed Dataset (RDD) in Spark, building machine learning model using PySpark.
Course Title: Natural Language Processing
Course Code: DS5601
L T P C: 3-0-0-3
Category: Elective UG/PG
Prerequisite Courses: Probability and Statistics, Machine Learning
Content: Introduction: Why NLP is hard, Why NLP is useful, Linguistics background; Words: Structure (Morphology, spell-checking), Semantics (Basic ideas in Lexical semantics, WordNet, word similarity), Word Sense Disambiguation, Distributional semantics (Latent Semantic Analysis, Word2Vec); Syntax: Linguistic aspects of syntax, Part-of-Speech tagging, Grammars, Parsing; Language Modelling: N-gram model, Smoothing, Probabilistic topic models, Neural Language Models (RNN, Transformer models), Contextualized Word Embeddings (BERT); Information Extraction (IE): Named Entity Recognition, Relation Extraction, Semantic Web; Semantics: Linguistic aspects of semantics, Logical Semantics, Semantic Role Labelling; Discourse and Pragmatics: Linguistic aspects of discourse and pragmatics, Reference resolution; NLP Applications: Summarization, Question-Answering, Sentiment analysis; Machine Translation: Rule based Machine Translation, Statistical Machine Translation, Neural Machine Translation; Natural Language Generation (NLG); Error analysis and evaluation of NLP Systems: Summarization, Machine Translation, NLG; Ethical and social aspects of NLP.
Course Title: Information Retrieval
Course Code: DS5603
L T P C: 3-0-0-3
Category: Elective UG/PG
Prerequisite Courses: Probability & Statistics, Programming
Content: Introduction to Information Retrieval - Basic Text Processing: Tokenization, Stopwords, Stemming, Lemmatization, Zipf's and Heap's law; Spelling correction and Edit distances: Hamming distance, Longest common Subsequence, Levenstein edit distance, Boolean Retrieval Model; Basic Ranking and Evaluation Measures - Vector Space Model, TF*IDF, IR Evaluation: Precision, Recall, F-measures, Mean Reciprocal Rank (MRR), Mean Average Precision (MAP), Normalized Discounted Cumulative Gain (NDCG), Designing test collection, relevance judgments; Probabilistic Retrieval Model - Introduction: Generative Model, Probabilistic Ranking Principle, Binary Independence Model, Okapi 25, Bayesian Networks for IR; Statistical Language Model - Basics of Language Model, Query-likelihood Approach and different Smoothing Methods, Advance Query Type: Query expansion, Relevance feedback, Novelty & Diversity; Topic Model - Introduction to topic model, Latent Semantic Indexing; Probabilistic Latent Semantic Indexing, Latent Dirichlet Allocation, Topic model for IR; Link Analysis - Introduction: World Wide Web as Graph, PageRank, HITS, Topic-specific and Personalized PageRank; Indexing and Searching - Different Compression Methods: Ziv-Lempel, Variable-Byte, Gamma, Golomb, Gap encoding, Query Processing: TAAT, DAAT, WAND, Fagin's algorithm, Near Duplicate Detection: Shingling, Min-wise independent permutations, locality sensitive hashing; Retrieval using unsupervised techniques - Retrieval using word-embeddings and Clustering; Retrieval using Supervised ML - Introduction to Learning to Rank for retrieval, Retrieval using classification.
Course Title: Responsible Artificial Intelligence
Course Code: DS6004
L T P C: 3-0-0-3
Category: Elective UG/PG
Prerequisite Courses: A course on Artificial Intelligence or Machine Learning or Deep Learning
Content: Artificial Intelligence Fundamentals; Introduction to responsible AI - Need for ethics in AI. AI for Society and Humanity; Fairness and Bias - Sources of Biases, Exploratory data analysis, limitation of a dataset, Preprocessing, inprocessing and postprocessing to remove bias, Group fairness and Individual fairness, Counterfactual fairness; Interpretability and explainability - Interpretability through simplification and visualization, Intrinsic interpretable methods, Post Hoc interpretability, Explainability through causality, Model agnostic Interpretation; Ethics and Accountability - Auditing AI models, fairness assessment, Principles for ethical practices; Privacy preservation - Attack models, Privacy-preserving Learning, Differential privacy, Federated learning; Case study (Any three) - Recommendation systems, Medical diagnosis, Hiring/ Education, Computer Vision, Natural Language Processing.
Course Title: Topics in Machine Learning
Course Code: DS5606
L T P C: 3-0-0-3
Category: Elective UG/PG
Prerequisite Courses: None
Content: Introduction to the course, recap of linear algebra (vector derivative) and probability theory (Bayes Rule) basics; Regression: linear regression, ridge regression (Probabilistic interpretation); Classifier - Linear classification (e.g. perceptron, maximum margin, logistic regression), Non linear classification (e.g. KNN, use of kernel in SVM); Evaluation and Model Selection: ROC Curves, Evaluation Measures, Cross validation, Significance tests; Ensemble Methods - Boosting, Bagging, Decision Trees, Random Forests; Feature extraction (Principal Component Analysis, Canonical Correlation Analysis); Clustering: K-mean, Gaussian Mixture Model, Expectation Maximization, density based clustering; Sequential Learning (HMM).
Course Title: Computer Vision
Course Code: DS5602
L T P C: 3-1-0-4
Category: Elective UG/PG
Prerequisite Courses: Basics of Machine Learning, Deep Learning
Content: Introduction to images, 2D geometric transformations, Geometric transformation estimation, Visual Features and Representations: Edge, Blobs, Corner Detection; Scale Space and Scale Selection; SIFT, SURF; HoG, LBP, etc. Visual Matching: Bag-of-words, VLAD; RANSAC, Hough transform; Pyramid Matching; Optical Flow; Introduction, Image classification, Loss function and optimization, Neural networks, CNN architectures: AlexNet, ZFNet, VGG, InceptionNets, ResNets, DenseNets, Training; Review of RNNs; CNN + RNN Models: Spatio-temporal Models, Action/Activity Recognition; CNNs for Detection: Background of Object Detection, R-CNN, Fast R-CNN, Faster R-CNN, YOLO, SSD, RetinaNet; CNNs for Segmentation: FCN, SegNet, U-Net, Mask-RCNN, Visualizing CNN features DeepDream, Style Transfer; Attention Models, Generative Models: GAN, VaE, Efficient hardware for deep learning, hyperspectral imaging.
Course Title: Topics in Deep Learning
Course Code: DS5607
L T P C: 3-0-0-3
Category: Elective UG/PG
Prerequisite Courses: None
Content: Revision of Linear Algebra, Probability, Optimization; Basics of learning: model, weight, data, loss functions, learning, prediction, output functions, connection with gradients and optimization, linear and non-linear classifiers; Perceptron - model, algorithm, basic neuron, XOR; Feed forward neural network (FNN) - Multi-layer perceptrons, weights, learning problem, Forward propagation, Backpropagation, Regularization methods, dropout, batchnorm, data augmentation; Convolutional neural network (CNN) - Convolution, filters, padding, pooling, CNN for image classification, Example architectures, various applications, Recurrent neural network (RNN) - Recurrence, various types of RNNs for different applications, Drawbacks of RNNs, Long Short Term Memory networks (LSTM), GRU, Attention and memory networks; Auto-encoders - Encoder-decoder, seq2seq case study, dimensionality reduction, denoising, Representation learning; Generative adversarial network (GAN), WGAN, conditional-GAN; Transformers.
Course Title: Advanced Machine Learning
Course Code: DS6001
L T P C: 3-0-0-3
Category: Elective UG/PG
Prerequisite Courses: None
Content: Revision of Machine Learning and Deep learning; Hierarchical Bayesian models: generative models, Topic models; Bayesian nonparametrics: Gaussian process, neural process, dirichlet process; Deep generative models: Generative adversarial network; Bayesian deep learning; Kernel Learning; Machine learning with graph and other structural data; Responsible machine learning: fairness, interpretability, causality, privacy; Advanced paradigms of learning: Multi-task learning, Transfer-learning, Multi-view-learning, Meta-learning.
Course Title: Probabilistic Graphical Models
Course Code: DS6003
L T P C: 3-0-0-3
Category: Elective UG/PG
Prerequisite Courses: Linear Algebra, Probability, Basics of Machine Learning, and Programming in Python
Content: Introduction and review of Probability and Graph theory concepts; Bayesian Networks– Representation, Independence properties, d-separation; Undirected Graphical Models – Parameterization, Factors, Gibbs distribution Markov networks, independence properties; Exact inference – variable elimination, message passing, belief update; Approximate inference – forward sampling, likelihood weighting and importance sampling, Gibbs sampling; Variational Inference; Learning graphical models – parameter estimation, maximum likelihood estimation for Bayesian networks, Bayesian parameter estimation; Sequence models – Hidden Markov Models – Representation, Inference, Viterbi decoding, Learning HMM – forward-backward algorithm; Advanced topics.
Course Title: Time Series Modeling and Analysis
Course Code: DS5608
L T P C: 3-0-0-3
Category: Elective UG/PG
Prerequisite Courses: Basics of Probability and Statistics, Linear Algebra, and familiarity with Python
Content: Introduction; Data mining and time series analysis, basic concepts in time series; Statistical Estimation; Standard distributions, basic concepts in statistical estimation, estimation of second-order properties; Transformation and Decomposition of Time Series; estimating trend, seasonal component; Quantifying Correlation Structure in Standard Models; Standard ARMA(p, q) models, Difference equations, Anatomy of MA(q), AR(p), and ARMA (p,q) models; Spectral Properties of standard models; Power spectrum Forecasting Methods; Optimal forecast – conditional expectation, Optimal linear forecast – Wiener’s approach, Forecasting using ARMA(p, q) models, recursive algorithms for forecasting; Estimation of Parameters in Standard models; Estimation in AR(p) models - Maximum likelihood and least squares approach, Estimation in MA(q) and ARMA(p, q) models, Methods of moments and its generalization; Partial Autocorrelation Function; Wold’s Decomposition; Deterministic and Nondeterministic component; Model Selection – Box and Jenkin’s Approach; Model selection criteria and examples; Volatility models; ARCH and GARCH models; State Space Models and Kalman Filtering; State space representation, Kalman filtering – estimation and prediction; Special Topics; Nonlinear models for time series analysis, long memory models, multivariate models.
Course Title: Mathematics for Data Science
Course Code: DS5004
L T P C: 3-0-0-3
Category: Core PG
Prerequisite Courses: Basics of Probability and Statistics, Linear Algebra
Content: Introduction- Importance of linear algebra, probability and statistics; vector spaces; linear span; linear independence; basis; dimension; linear maps and matrices; change of basis; fundamental subspaces of a matrix; fundamental theorem of linear algebra; Systems of linear equations; LU decomposition; introduction to vector calculus; least-squares method; Norms; inner products; Cauchy-Schwarz inequality; orthonormal basis; orthogonal projections; Gram-Schmidt orthogonalization; QR decomposition; Eigenvalues; eigenvectors; Hermitian matrices; Cholesky decomposition; diagonalization; spectral theorem; principal component analysis; singular value decomposition; low-rank matrix approximation; Eckart-Young theorem; Probability axioms; conditional probability; independence; Bayes’ theorem; expectation; variance; Bernoulli distribution; binomial distribution; multinomial distribution; Dirichlet distribution; change of variables; Gaussian distribution (univariate and multivariate); conditional and marginal Gaussian distribution; Law of large numbers; central limit theorem; concentration inequalities (Markov, Chebyshev, Hoeffding, Chernoff); Conjugate prior; noninformative priors; sufficient statistic; exponential family; Sampling distributions; chi-squared and student-t distribution; Maximum likelihood estimation; least-squares method revisited; interval estimation; confidence intervals; hypothesis testing; Correlation functions
Course Title: Machine Learning
Course Code: DS5006
L T P C: 3-0-2-4
Category: Core PG, Elective UG
Prerequisite Courses: Basics of Machine Learning, Programming
Content: Introduction and review of basic pre-requisites – Linear Algebra, Probability Theory; Basics of supervised Learning: Hypothesis learning, Empirical Error, and Generalization Error, Empirical Risk Minimization.; Tree-based Learning: Decision Trees and Random Forest; Linear Regression: Closed-form Solution, Convex Loss Function, Gradient Descent, Probabilistic Interpretation – Maximum-Likelihood Estimation, Bias-Variance Analysis, Regularization – Maximum Aposteriori Estimation - Ridge and LASSO; Linear Classification: Regressor as a Classifier, Linear Discriminant Function, Logistic Regression, Second-order Optimization Method -Newton Raphson Method; Kernel Methods: Margin, Maximum Margin Classifiers, Support Vector Machine (SVM), Constrained Optimization, Lagrange Multipliers, KKT Conditions, Primal and Dual Formulation, Soft-margin SVM, Sequential Minimization Optimization, Kernel Functions and the Kernel Trick, Support Vector Regression; Experimental Design and Analysis: Model Selection, Cross Validation, Performance Measures, Hypothesis Testing, Confidence Interval Estimation; Basics of unsupervised learning: k-Means Clustering, Gaussian Mixture Models; Dimensionality Reduction: Principal Component Analysis and its Variants, Linear Discriminant Analysis
Course Title: Deep Learning
Course Code: DS5007
L T P C: 3-0-2-4
Category: Core PG, Elective UG
Prerequisite Courses: Basics of Machine Learning, Programming
Content: Revision of Linear Algebra, Probability, Optimization. Basics of learning: model, weight, data, loss functions, learning, prediction, output functions, connection with gradients and optimization, linear and non-linear classifiers; Perceptron: model, algorithm, basic neuron, XOR; Feed forward neural network; Multi-layer perceptrons (MLP), weights, learning problem, Forward propagation; Activation functions, non-linearity, revisiting XOR; MLP as universal function approximators; Backpropagation; Complete derivation of the backprop algorithm; Implementation of the backprop algorithm; Evaluation methods, Generalization, and Regularization methods, dropout, batch normalization, data augmentation; Optimization techniques in deep learning: stochastic gradient descent (SGD), batch mode, momentum methods (Adam); Convolutional neural network (CNN); Convolution, filters, padding, pooling; CNN for image classification; Key architectures of CNNs modeling choice and evolution; Backpropagation; Understanding CNN learning through Grad-Cam and Deconvolution; Recurrent neural network (RNN) - Recurrence, various types of RNNs for different applications; Drawbacks of RNNs, Long Short Term Memory networks (LSTM), GRU; Backpropagation through time; Auto-encoders- Encoder-decoder, seq2seq case study, dimensionality reduction, denoising, Representation learning; Attention and Transformers; Deep generative models- Generative adversarial network (GAN), WGAN, conditional-GAN; Variational auto-encoders (VAE), Flow Models, Diffusion Models; Hopfield networks, Boltzmann machine, Restricted Boltzmann machine
Course Title: Data Mining
Course Code: DS5612
L T P C: 3-0-0-3
Category: Elective UG/PG
Prerequisite Courses: Basics of Linear Algebra, Probability, Programming
Content: Introduction: Life cycle of data and ETL pipeline; Visualization: Techniques for visualizing numerical, non-numerical data, geospatial and multi-dimensional data, Visual Analytics, introduction to Psychophysics, Grammar of Graphics; Pattern mining: market-basket model, frequent itemsets, association rules, Apriori and PCY algorithms, mining sequential patterns; Finding similar items: near-neighbor search, Document Shingling, Minhash, locality sensitive hashing; Classification models: Decision trees, kNN, Naive Bayes; Ensemble learning such as Random Forests, AdaBoost, XGBoost; Model comparison, Metrics for evaluating classifiers, cross validation; Clustering algorithms: basic approaches such as partitioning, hierarchical, grid-based, density-based; advanced topics such as probabilistic clustering, high-dimensional clustering, bi-clustering and correlation clustering; Dimensionality reduction techniques (PCA, SVD and MDS); Recommender systems: Collaborative Filtering, Matrix Factorization (NMF) based approaches; Mining social networks: graph clustering and partitioning, community discovery, centrality measures; Data Mining in the wild: Sampling, resampling and bootstrapping, Monte Carlo simulations, hypothesis and multiple testing, multi-armed bandits, A/B and bucket testing; Presenting data mining results: Narrative Visualization and storytelling, errors and uncertainty
Course Title: Data Engineering
Course Code: DS5003A
L T P C: 1-0-3-3
Category: Core PG
Prerequisite Courses: None
Content: Data Structures and Algorithms - Data structures: array, matrix, stack, queue, dictionary, graph; Basics of algorithm, pseudo-code, testing-debugging; Searching, sorting; Graph traversal: depth-first, breadth-first; Trees, Binary search; Problems related to data science, Complexity analysis Big O; Database Management Systems - Basics of databases, ER diagram, RDBMS basics; SQL: basics, aggregation, Multi-table queries: joins, Views, functions, procedures, triggers, MongoDB; Data-dependent systems - basic architecture, developmental principles, and software engineering; Building front-end GUI and connecting with back-end database
Course Title: Business Analytics
Course Code: DS5610
L T P C: 3-0-0-3
Category: Elective UG/PG
Prerequisite Courses: Machine Learning
Content: Introduction to Business Analytics - An overview and Data Science interface; Introduction to Business Decision Making - familiarity with business functions: Demonstrate business situations that warrant the application of analytics; Descriptive Analytics - Data exploration and confirmation - Use of hypothesis testing. clustering, factor analysis, and other techniques. Business Functions with case analysis from different functions such as Operations; Quality, Supply Chain, etc. Predictive Analytics – Use of regression and classification in business decision making; Marketing mix model, healthcare analytics, fraud analytics, HR analytics; Market Basket Analysis, Customer Lifetime Value; Prescriptive Analytics - Demand forecasting and optimization
Mehta Family School of Data Science & Artificial Intelligence
Department of Data Science
Indian Institute of Technology Palakkad
Kanjikkode | Palakkad
Kerala | 678623
+91 491 2091201
mfsdsai@iitpkd.ac.in
Designed and Developed by Alakesh Bezbaruah
Maintained by Narayanan C Krishnan, Nikhil Krishnan M, and Swapnil Hingmire