VectorSpaceIR

An Information Retreival system based on the Vector space model. Returns items based on cosine similarity score. Written in Java 8. To use download and unzip, place files in the "/src/Files/" directory.

Uses the Porter stemmer.

Issues:

  • Currently hardcoded to have maximum 15 files which needs to be fixed.
  • Uses Double type vs double.
  • Needs refactoring, better use of data structures.