Skip to content

interactive_clustering

  • Name: cognitivefactory.interactive_clustering
  • Description: Python package used to apply NLP interactive clustering methods.
  • Author: Erwan SCHILD
  • Created: 17/03/2021
  • Licence: CeCILL (https://cecill.info/licences.fr.html)

Three modules are available:

  • constraints: it provides a constraints manager, that stores annotated constraints on data and gives some feedback on information deduced (such as the transitivity between constraints or the situation of inconsistency). See interactive_clustering/constraints documentation ;
  • sampling: it provides several constraints sampling algorithm, that selecte relevant contraints to annotate by an expert. See interactive_clustering/sampling documentation ;
  • clustering: it provides several constrained clustering algorithms, that partition the data according to annotated constraints. See interactive_clustering/clustering documentation ;
  • utils: it provides several basic functionnalities, like data preprocessing and data vectorization. See interactive_clustering/utils documentation.