M.Tech. Data Science and Engineering

First Semester

• Mathematical foundations for Data Science

• Introduction to Data Science

• Computer Organization and Software Systems

• Data Structures and Algorithm Design

Second Semester

• Introduction to Statistical Methods

• Elective - I

• Elective - II

• Elective - III

Third Semester

• Big Data Systems

• Elective - IV

• Elective - V

• Elective - VI

Fourth Semester

• Dissertation

Electives

• Data Warehousing

• Graphs – Algorithms and Mining

• Deep Learning

• Probabilistic Graphical Models

• Ethics for Data Science

• Optimization Techniques for Analytics

• Data Management for Machine Learning

• Natural Language Processing

• Design of Experiments for Data Science

• Information Retrieval

• Data Visualization and Interpretation

• Stream Processing and Analytics

• Artificial and Computational Intelligence

• Machine Learning #*

• Applied Machine Learning

Course Wise Syllabus

1. Mathematical Foundations for Data Science

• Vector and matrix algebra, systems of linear algebraic equations and their solutions; Eigenvalues, eigenvectors and diagonalization of matrices; multivariate calculus, vector calculus, Jacobian and Hessian, multivariate Taylor series, gradient descent, unconstrained optimization, constrained optimization, nonlinear optimization, stochastic gradient descent, dimensionality reduction and PCA, optimization for support vector machines.

2.Introduction to Data Sciencee

• Data Analytics, Data and Data Models, Data wrangling, Feature Engineering, Classification and Prediction, Association Analysis, Clustering, Anomaly Detection, exploratory / explanatory data analysis with visual storytelling, Ethics for Data Science.

3.Computer Organization and Software Systems

• Computer organization, architecture aspects and operating system concepts

• Advanced systems and techniques used for data processing

• Introduction to computer organization, architecture; Introduction to operating systems; ISA - RISC etc. Processes and Threads; Scheduling, Concurrency; Memory Management

4. Data Structures and Algorithm Design

• Applications of basic and advanced data structures & algorithms

• How to determine the space and time complexities of various algorithms

• Identifying and choosing the relevant data structures and algorithms for a given problem and justifying the time and space complexities involved

• Stack, Queue, Hash Tables, Graphs, Bloom Filters, Trees, Sets, Dictionary, Dynamic Graphs; Divide-and-conquer, Dynamic Programming, Graph Algorithms, Greedy Algorithms - Spanning Tree, Amortized Analysis, Huffman encoding; Page Rank, Map Reduce (map, fork, join etc); Complexity analysis

5. Introduction to Statistical Methods

• Basic probability concepts, Conditional probability, Bayes Theorem, Probability distributions, Continuous and discrete distributions, Transformation of random variables, estimating mean, variance, covariance, Hypothesis Testing, Maximum likelihood, ANOVA – single factor, dual factor, time series analysis: AR, MA, ARIMA, SARIMA, sampling based on distribution, statistical significance, Gaussian Mixture Model, Expectation Maximization.

6.Big Data Systems

• Concepts related to big data and its processing

• Applying the concepts of storage, retrieval, interfaces and processing frameworks to a given problem and design solutions for the same by choosing the relevant ones

• What is big data - are existing systems sufficient; Data Warehouse v/s Data Lakes; Hadoop - Components; Storage - Relational DBs/ NoSQL dbs / HDFS / HBase / Object Data stores - S3; Serialization; Interfaces - Hive/ Pig; Stream Processing; Spark; Mahout Advanced Topics in Data Processing

7. Data Management for Machine Learning

• Data Models and Query Languages: Relational, Object-Relational, NoSQL data models; Declarative (SQL) and Imperative (Map Reduce) Querying; Data Encoding: Evolution, Formats, Models of dataflow; Machine learning workflow; Data management challenges in ML workflow; Data Pipelines and patterns; Data Pipeline Stages: Data extraction, ingestion, cleaning, wrangling, versioning, transformation, exploration, feature management; Modern Data Infrastructure: Diverse data sources, Cloud data warehouses and lakes, Data Ingestion tools, Data transformation and modelling tools, Workflow orchestration platforms; ML model metadata and Registry, ML Observability, Data privacy and anonymity.

8.Design of Experiments for Data Science

• Introduction and importance of Experimental Design, Testing of Hypothesis, Designs with One Source of Variation, Multiple Comparison Testing, Interaction Effect, Factorial Experiment, Fractional Factorial Designs & Confounding, Latin Squares and Graeco-Latin Squares, Fractional-Factorial Designs, Taguchi Design, Designs with Random Effects, Optimal Designs and Model Uncertainty, Design for Nonlinear Model, Sequential Designs.

9. Ethics for Data Science

• Nature of data - data as a by-product of computing, operations data (e.g., sales/marketing), surveillance data (business or government), data collected for research; Ethics - What are ethics, need for ethics, Ethical concerns in computing and analytics; Why data science needs ethics?; Issues -political/social, liberty and justice, fairness and equality, business competitiveness, privacy, anonymity, and security; Data Ownership, Informed Consent, Security Risks (Privacy, Anonymity, Integrity, and Provenance); Ethical methods for sourcing/collecting data, and for storage/ distribution of data. Data validation. Algorithmic Fairness and Case Studies; Solutions to address ethical issues for government, corporations/organizations, research, public use of data, social norms, legal compliance, and case studies. Data ethics in specific domains - e.g. health care, finance, and social studies/research.

10.Optimization Techniques for Analytics

• Role of optimization in different types of analytics, Introduction to Linear Programming, LP Model and graphical solution, Primal Simplex method, Dual Simplex and Post Optimality Analysis, Revised Simplex method with examples, Application of linear programming in transportation, assignment problems, Integer linear programming, mixed integer programming, complexity analysis, branch and bound techniques, goal programming, Network models - critical path method and PERT, Dynamic programming, game theory, additional meta heuristic techniques, 2-3 case studies from relevant industry domains.

11.Natural Language Processing

• Natural Language Understanding and Generation, N-gram and Neural Language Models, Word to Vectors / Word Embedding (Skip gram/CBOW, Glove, BERT/ XLM, MURIL), Part of Speech Tagging, Hidden Markov Models, Parsing - Syntactic, Statistical, Dependency, Word Sense Disambiguation, Semantic Web Ontology.

Choice of Electives is made available to enrolled students at the beginning of each semester. Students' choice will be taken as one of the factors while deciding on the Electives offered. However, Electives finally offered will be at the discretion of the Institute.