Could you recommend libraries to search sentences which include semantically similar phrases to the query sentence's?

More semantically common phrases included, more scores.

I think I should be able to represent a sentence with semantic components and their relations.

To solve the problem, I thought two options.

  1. To index paraphrases as normalized forms with the elasticsearch.
  2. To make an DNN model to represent sentences with feature vectors.

I'm studying following fields

  1. https://aclweb.org/aclwiki/index.php?title=Paraphrase_Identification_(State_of_the_art)
  2. http://meaningtext.net/wiki/Paraphrase

I found following related libraries. Most of them are made for the question answering problem.

  1. github.com/ProjetPP/PPP-QuestionParsing-Grammatical
  2. uima.apache.org/
  3. deepdive.stanford.edu/
  4. opencog.org/
  5. oaqa.github.io/
  6. aksw.org/Projects/openQA.html
  7. ailao.eu/yodaqa/
  8. knowitall.cs.washington.edu/paralex/

I haven't used any libraries above yet.

I'm focusing on implementation.

Please recommend useful resources such as libraries, articles, and keywords to survey. Reviews of resources will be very helpful, too.

  • $\begingroup$ I doubt you will find this ready-made in a library. However I did recently hear about some technologies that might help you, let me see if I can dig up the conversation I had $\endgroup$ – shadowtalker Sep 4 '16 at 12:23

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.