2CSE50E18: Information Retrieval

Learning Outcomes: 
After learning the course the students should be able to
Explain the concepts of indexing, vocabulary, normalization and dictionary in Information Retrieval
Define a boolean model and a vector space model, and explain the differences between them
Explain the differences between classification and clustering
Discuss the differences between different classification and clustering methods
Choose a suitable classification or clustering method depending on the problem constraints at hand
Implement classification in a boolean model and a vector space model
Implement a basic clustering method
Give account of a basic spectral method
Evaluate information retrieval algorithms, and give an account of the difficulties of evaluation
Explain the basics of XML and Web search.
Unit NoTopics

Basics of Information Retrieval and Introduction to Search Engines; Boolean Retrieval-: Boolean queries, Building simple indexes, Processing Boolean queries

Term Vocabulary and Posting Lists

Choosing document units, Selection of terms, Stop word elimination, Stemming and lemmatization, Skip lists, Positional postings and Phrase queries; Dictionaries and Tolerant Retrieval: Data structures for dictionaries, Wildcard queries, Permuterm and K-gram indexes, Spelling correction, Phonetic correction

Index Construction

Single pass scheme, Distributed indexing, Map Reduce, Dynamic indexing; Index Compression - Statistical properties of terms, Zipf's law, Heap's law, Dictionary compression, Postings file compression, Variable byte codes, Gamma codes

Vector Space Model

Parametric and zone indexes, Learning weights, Term frequency and weighting, Tf-Idf weighting, Vector space model for scoring, variant tf-idf function.

Computing Scores in a Complete Search System

Efficient scoring and ranking, Inexact retrieval, Champion lists, Impact ordering, Cluster pruning, Tiered indexes, Query term proximity, Vector  space scoring and query operations

Evaluation in Information Retrieval

Standard test collections, unranked retrieval sets, Ranked retrieval results, Assessing relevance, User utility, Precision and Recall, Relevance feedback, Rocchio algorithm, Probabilistic relevance feedback, Evaluation of relevance feedback

Probabilistic Information Retrieval

Review of basic probability theory, Probability ranking principle, Binary independence model, Probability estimates, probabilistic approaches to relevance feedback. Text Classification- Rocchio classifier, KNearest  neighbor classifier, Linear and nonlinear classifiers, Bias-variance tradeoff, Naïve Bayes and Support Vector machine basedclassifiers

Text Clustering

Clustering in information retrieval, Evaluation of clustering, KMeans and Hierarchical clustering. Introduction to Linear Algebra, Latent Semantic Indexing

Text Books: 
Name : 
An Introduction to Information Retrieval
C. D. Manning, P. Raghavan, and H. Schutze
Cambridge University Press, 2009.
Reference Books: 
Modern Information Retrieval
R. Baeza-Yates and B. Ribeiro-Neto
Pearson Education, 1999.
Syllabus PDF: 
PDF icon ELECTIVE III (IR) .pdf105.43 KB