https://github.com/kenoir/polysemous

Some old university work in Prolog on word sense disambiguation
https://github.com/kenoir/polysemous

Last synced: 9 days ago
JSON representation

Some old university work in Prolog on word sense disambiguation

• Host: GitHub
• URL: https://github.com/kenoir/polysemous
• Owner: kenoir
• Created: 2012-12-31T17:18:30.000Z (about 12 years ago)
• Default Branch: master
• Last Pushed: 2013-02-15T16:47:07.000Z (almost 12 years ago)
• Last Synced: 2024-11-08T21:59:30.393Z (2 months ago)
• Language: Perl
• Size: 15.7 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
polysemous

==========

Some old university work in Prolog on word sense disambiguation.

Shoving it here for safekeeping and reference for anyone who

wants to see how not to write good Prolog.

Here be dragons!

In Summary

----------

Polysemy causes inaccuracy in information retrieval by allowing

a word to have more than one meaning, for example: 

    Dog: someone who is morally reprehensible.

    Dog: a member of the genus Canis.

Polysemy causes the results of a search to be inaccurate by 

including all the documents containing the unwanted senses of 

a word.

    I dig that groovy cat!

This project is an attempt to increase the accuracy of such 

queries by reducing the problems associated with polysemy by 

identifying the meaning of each word in a document (a process 

called sense tagging) and using those senses in place of words 

to search for a document.

We attempt to achieve this by making use of the [Lesk algorithm](http://en.wikipedia.org/wiki/Lesk_algorithm), 

the [Vector Space Model](http://en.wikipedia.org/wiki/Vector_space_model), and the [Wordnet project](http://wordnet.princeton.edu/).

This project takes a shallow approach to the problem of [Word Sense Disambiguation](http://en.wikipedia.org/wiki/Word-sense_disambiguation).