Supervised feature extraction for text categorization
Abstract
This paper concerns finding the `optimal' number of (non-overlapping) word groups for text classification. We present a method to select which words to cluster in word groups and how many such word groups to use on the basis of a set of pre-classified texts. The method involves a `greedy' search through the space of possible word groups. The words are grouped according to the `Jensen-Shannon divergence' between the corresponding distributions over the classes. The criterion to decide which number of word groups to use is based on Rissanen's MDL Principle. We present empirical results that indicate that the proposed method performs well. Furthermore, the proposed method outperforms cross-validation in the sense that far fewer word groups are selected while prediction accuracy is just slightly worse. For the experimentation we used a subset of the `20 Newsgroup' data set.
Fichier principal
verbeek00bnl.pdf (206.76 Ko)
Télécharger le fichier
Ver00a.png (63.98 Ko)
Télécharger le fichier
Origin : Files produced by the author(s)
Format : Figure, Image