Compari is a comparison engine for named-entities.
In particular, this is a project to investigate the how Wikipedia's graph structure can be used to assist in the discovery and ranking of different topics. It will be used for my CS 221 (Artificial Intelligence) final project at Stanford.
Poke around
src/java
contains code for parsing wikipedia and performing Latent Dirichlet Allocation (LDA)src/lda/core.clj
contains all the stuff needed compute Topic-Sensitive PageRank on Wikipediasrc/query/core.clj
query engine codesrc/query/handler.clj
web interface for query engine
Copyright © 2013 Thomas Dimson. All rights reserved.