Enhancing Text Classification by Information Embedded in the Test Set

  • Gabriela Ramírez-de-la-Rosa
  • Manuel Montes-y-Gómez
  • Luis Villaseńor-Pineda
Conference paper

DOI: 10.1007/978-3-642-12116-6_53

Part of the Lecture Notes in Computer Science book series (LNCS, volume 6008)
Cite this paper as:
Ramírez-de-la-Rosa G., Montes-y-Gómez M., Villaseńor-Pineda L. (2010) Enhancing Text Classification by Information Embedded in the Test Set. In: Gelbukh A. (eds) Computational Linguistics and Intelligent Text Processing. CICLing 2010. Lecture Notes in Computer Science, vol 6008. Springer, Berlin, Heidelberg

Abstract

Current text classification methods are mostly based on a supervised approach, which require a large number of examples to build models accurate. Unfortunately, in several tasks training sets are extremely small and their generation is very expensive. In order to tackle this problem in this paper we propose a new text classification method that takes advantage of the information embedded in the own test set. This method is supported on the idea that similar documents must belong to the same category. Particularly, it classifies the documents by considering not only their own content but also information about the assigned category to other similar documents from the same test set. Experimental results in four data sets of different sizes are encouraging. They indicate that the proposed method is appropriate to be used with small training sets, where it could significantly outperform the results from traditional approaches such as Naive Bayes and Support Vector Machines.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Gabriela Ramírez-de-la-Rosa
    • 1
  • Manuel Montes-y-Gómez
    • 1
  • Luis Villaseńor-Pineda
    • 1
  1. 1.Laboratory of Language TechnologiesNational Institute of Astrophysics, Optics and ElectronicsPue.Mexico

Personalised recommendations