Nonnegative factor analysis for text document clustering


Skovajsova L., Mokris I.

9th WSEAS International Conference on Simulation, Modelling and Optimization, Budapest, Hungary, 3 - 05 September 2009, pp.345-346 identifier

  • Publication Type: Conference Paper / Full Text
  • City: Budapest
  • Country: Hungary
  • Page Numbers: pp.345-346

Abstract

This paper deals with text document clustering by means of neural network used for preprocessing and next, the nonnegative factor analysis is applied to create certain amount of clusters. The results on the part of Reuters-21578 collection show that the given number of clusters is created, and the difference between clusters is counted as the cosine similarity between centroids of the particular clusters. Results show that if the data are preprocessed by PCA, the non-negative factor analysis divides documents into given number of clusters quite successfully.