Information Retrieval

Published

November 1, 2024

Information retrieval, also abbreviated IR, is the task of finding (or retrieving) text documents that contain some desired information (e.g., the answer to a user’s search query) from a large collection of documents.

Lecture

Demo Notebook

Lab

In this lab, you will apply basic techniques from information retrieval to implement the core of a minimalistic search engine.

Reading Material

We recommend the following book, which has free online edition:

The material in this lecture focuses on the following parts:

  • Sections 2.1–2.2 (tokenization and preprocessing steps)
  • Chapter 6 (tf–idf and the vector space model)
  • Sections 8.1–8.4 (evaluation metrics)