In this paper, we present a new feature selection method based on document frequencies and statistical values. We also present a new similarity measure to calculate the degree of similarity between documents. Based on the proposed feature selection method and the proposed similarity measure between documents, we present three methods for dealing with the Reuters-21578 top 10 categories text categorization. The proposed methods get higher performance for dealing with the Reuters-21578 top 10 categories text categorization than that of the method presented in [4]. © Springer-Verlag Berlin Heidelberg 2006.
CITATION STYLE
Lee, L. W., & Chen, S. M. (2006). New methods for text categorization based on a new feature selection method and a new similarity measure between documents. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4031 LNAI, pp. 1280–1289). Springer Verlag. https://doi.org/10.1007/11779568_135
Mendeley helps you to discover research relevant for your work.