Information Retrieval (IR)

News and Updates:

Topics (Tentative):

Introduction, DB vs IR, Tokenization, Indexing, Representing terms and documents, Scoring models, Query-independent ranking, Learning to Rank, Evaluation in IR, Advanced Topics from Research Papers

Pre-requisites:

Understanding of Probability, satistics, basic machine learning concepts

Grading Scheme (Tentative):

Quiz: 20%, End Term: 30-40%, Project: 40-50%

Reference Material:


Lecture Schedule
Date/Week Topic Comments
Week 1-2
  • Introduction to IR, tokenization, text normalization, indexing
Week 3-4
  • Vector space model and considerations
  • Hands on session on vectorization
  • Project announcement
Week 5-6
  • Cosine similarity, pivoted document length normalization
  • Language modeling
Week 7-8
  • Evaluation in IR
  • Introduction to word representations
  • Hands on session on word representations
  • Mid-term project presentations
Week 9-10
  • Probabilistic IR
  • Latent Semantic Analysis, Learning to Rank
Week 11-12
  • Query agnostic scoring: Pagerank, HITS
  • Introduction to Neural IR
Week 13-15
  • Search Result Diversification
  • Invited Talk
  • Final Project Presentations