Support vector machines based Arabic language text classification system: feature selection comparative study

  • Authors:
  • Abdelwadood Mesleh

  • Affiliations:
  • Computer Engineering Department, Balqa' Applied University, Faculty of Engineering Technology, Amman, Jordan

  • Venue:
  • MATH'07 Proceedings of the 12th WSEAS International Conference on Applied Mathematics
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Feature selection is essential for effective and accurate text classification systems. This paper investigates the effectiveness of six commonly used feature selection methods, Evaluation used an in-house collected Arabic text classification corpus, and classification is based on Support Vector Machine Classifier. The experimental results are presented in terms of precision, recall and Macroaveraged F1 measure.