Home

pyConText Goals

collaborative development and improvement

• Github: https://github.com/chapmanbe/pyConTextNLP
  • forking
  • merging

algorithmic improvements

• interaction with termination terms
• modularity
• target identification
  • can either mark targets or read them in

document-level classification

* in RadNLP (not pyConText)
* rule language format

read in ontology files

* simple text?
* RDF?
* SWRL?

Functional pyConTextNLP

unit testing

* public corpora 
* MTSamples?
* 120 PITT sentences?

efficiency

* speed it up
* fastConText code (Jianlin)
* ranking modifiers according to utility
* parallel processing

visualization and customization

* notebooks