I am looking for any java library that calculates the semantic distance between two documents.

I know that WS4J can calculate semantic distances using wikipedia data, also lingpipe has some NLP functions, so does Lucene, so does Stanford NLP and also Apache OpenNLP

However, most of them has primitives or can calculate distances between words. What I need here is a way to compute semantic distances between documents OOTB. Java library.

  • About what length, how many thousand/million characters? – Nicolas Raoul Jun 13 '16 at 3:51
  • yes, we can assume US-en UTF8 documents. thanks – Leo Jun 13 '16 at 14:23
  • I found SEMILAR: A Semantic Similarity Toolkit. It is a Java library as well as standalone application. This library can compute similarity (I assume that is semantic distance) between two texts (documents). You may check it out here. *I recommend this on comment section rather than answer section because I have never tried using this library before). – meisyal Jun 19 '16 at 13:02
  • yep, this is pretty much what I am looking for. I am gonna give it a try. thx – Leo Jun 20 '16 at 12:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.