#IR class project - ranked retrieval using vector space model

• Author: Hariram S
• Date: 2016-09-27
• Version: 1.0.0
• Dataset: TUAW - Social Computing Data Repository
• Data source: http://socialcomputing.asu.edu/datasets/TUAW

##Requirements

Tested using:

• python v3.5.2
• porterstemmer.py source: https://tartarus.org/martin/PorterStemmer/
• nltk v3.2.1
• nltk data:
  • punkt (Punkt Tokenizer Models)
  • stopwords (Stopwords Corpus)

##Usage

1. Run 2to3 porterstemmer.py to make the stemmer compatible with python 3 (if not done already)
2. Run build_index.py to construct an inverted index from the data.
3. Run search.py "your query" [ k ] to get top k results (default k = 10).