https://github.com/sobotics/unistack

An experimental duplicate question search algorithm for Stack Exchange.
https://github.com/sobotics/unistack

Last synced: 10 months ago
JSON representation

An experimental duplicate question search algorithm for Stack Exchange.

• Host: GitHub
• URL: https://github.com/sobotics/unistack
• Owner: SOBotics
• License: wtfpl
• Created: 2016-05-02T10:00:14.000Z (about 10 years ago)
• Default Branch: master
• Last Pushed: 2018-04-21T14:27:35.000Z (about 8 years ago)
• Last Synced: 2025-07-04T18:45:53.766Z (11 months ago)
• Language: C#
• Homepage:
• Size: 446 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# UniStack

An experimental duplicate question search algorithm for Stack Exchange.

# Current Algorithm

*As this is an experimental project, everything is subject to change.*

Extract question sentence + context words, then model remianing words.

 - Question sentence: words starting from an MD or Wxx tagged word till the end of the sentence. 

 - Context words: any NN tagged word.

 - Model: PoS tagged bag-of-words (unigram).

 - Weighting: TF-IDF.

 - Similarity function: cosine.