TTDS Labs 2022/23 - A Search Engine Course

Lab 5: Query expansion for search engine queries using Pseudo Relevance Feedback

Retrieving the top 5 ranked terms from the top ranked document for each search query, based on the TFIDF formula, for an improved expanded search query.

Implementation Overview

1. Pre-processing the given text collection and building an inverted index for it.
2. Parsing the already created Ranked Information Retrieval results (results.ranked.txt) for each of the given queries (queries.lab3.txt), which are based on the TFIDF formula.
3. Taking the highest ranked document for each query and calculating the TFIDF for each term in that pre-processed document.
4. Retrieving the top 5 terms which are the most relevant terms to expand the original query with, in this case.
5. Parsing and pre-processing the given queries and writing everything to the results file Qm.1.5.txt.

Instructions taken from

https://www.inf.ed.ac.uk/teaching/courses/tts/labs/lab5.html