A basic search engine written in Java after taking the course Information Retrieval at UvA.
When taking the course Information Retrieval at the University of Amsterdam we built a sophisticated search engine in Python, using several frameworks. I wanted to see if I could do the same myself in Java, using no external libraries.
At the moment the search engine reads and tokenizes a (very small, handmade) set of documents, and ranks them using tf-idf scoring.
Future enhancements could be to build a general Score class, which can be extended to use other measures such as BM25, Dirichlet Prior, Jelinek-Mercer smoothing and even Learning-To-Rank at some point. The data set should also be significantly increased, as it currently only holds a very small sample of documents used for testing.