This project implements BM25 algorithm described in this paper for ranking documents according to relevance.
Make sure to run the setup.sh script. It will install all required dependencies.
Use python sample.py to see it in action. It may take some time, but eventually it will print to the console retrieved documents matching the query, sorted by relevance.
This implementation is best fit for small datasets because it lacks the inverted index required for fast querying in big datasets