https://github.com/jcla1/wikipedia_analyser

https://github.com/jcla1/wikipedia_analyser

Last synced: about 1 month ago
JSON representation

• Host: GitHub
• URL: https://github.com/jcla1/wikipedia_analyser
• Owner: jcla1
• Created: 2013-08-06T14:30:40.000Z (over 11 years ago)
• Default Branch: master
• Last Pushed: 2013-09-13T14:56:16.000Z (over 11 years ago)
• Last Synced: 2024-04-14T15:48:43.432Z (9 months ago)
• Language: Go
• Homepage:
• Size: 156 KB
• Stars: 7
• Watchers: 5
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Wikipedia analyser

Project for #YRS2013

Uses neural network from [jcla1/nn](https://github.com/jcla1/nn)

Basically analyses wikipedia lexically and structurally to find things that need to be improved.

Uses a (self-written) neural-net for analysing structure and an ngram model to analyse the sentence fragments.

Built around a pipeline that enables it to be distributed across multiple CPU cores.

To get started look at the different parts the project is divided into in [parts.go](parts.go)

I haven't tested the ngram code yet since I haven't got enough disk space to process the ngrams.