Take a look at the new version of this page: [ beta version ]. Tell us what you think.

ABSTRACT

Information Gain is a well-known and empirically proven method for high-dimensional feature selection. We found that it and other existing methods failed to produce good results on an industrial text classification problem. On investigating the root cause, we find that a large class of feature scoring methods suffers a pitfall: they can be blinded by a surplus of strongly predictive features for some classes, while largely ignoring features needed to discriminate difficult classes. In this paper we demonstrate this pitfall hurts performance even for a relatively uniform text classification task. Based on this understanding, we present solutions inspired by round-robin scheduling that avoid this pitfall, without resorting to costly wrapper methods. Empirical evaluation on 19 datasets shows substantial improvements.

REFERENCES

• 1: George Forman, An extensive empirical study of feature selection metrics for text classification, The Journal of Machine Learning Research, 3, 3/1/2003
• 2: Isabelle Guyon , Jason Weston , Stephen Barnhill , Vladimir Vapnik, Gene Selection for Cancer Classification using Support Vector Machines, Machine Learning, v.46 n.1-3, p.389-422, 2002 [doi>10.1023/A:1012487302797]
• 3: Mark A. Hall , Geoffrey Holmes, Benchmarking Attribute Selection Techniques for Discrete Class Data Mining, IEEE Transactions on Knowledge and Data Engineering, v.15 n.6, p.1437-1447, November 2003 [doi>10.1109/TKDE.2003.1245283]
• 4: Eui-Hong Han , George Karypis, Centroid-Based Document Classification: Analysis and Experimental Results, Proceedings of the 4th European Conference on Principles of Data Mining and Knowledge Discovery, p.424-431, September 13-16, 2000
• 5: Thorsten Joachims, Text Categorization with Suport Vector Machines: Learning with Many Relevant Features, Proceedings of the 10th European Conference on Machine Learning, p.137-142, April 21-23, 1998
• 6: Ron Kohavi , George H. John, Wrappers for feature subset selection, Artificial Intelligence, v.97 n.1-2, p.273-324, Dec. 1997 [doi>10.1016/S0004-3702(97)00043-X]
• 7: Dunja Mladenic , Marko Grobelnik, Feature Selection for Unbalanced Class Distribution and Naive Bayes, Proceedings of the Sixteenth International Conference on Machine Learning, p.258-267, June 27-30, 1999
• 8: Mladenic, D., & Grobelnik, M. (1998). Word sequences as features in text learning. 17th Electrotechnical and Computer Science Conference.
• 9: Rennie, J. (2001). Improving multi-class text classification with naive Bayes. Master's thesis, Massachusetts Institute of Technology (ch. 5).
• 10: Ian H. Witten , Eibe Frank, Data mining: practical machine learning tools and techniques with Java implementations, Morgan Kaufmann Publishers Inc., San Francisco, CA, 2000
• 11: Yiming Yang , Xin Liu, A re-examination of text categorization methods, Proceedings of the 22nd annual international ACM SIGIR conference on Research and development in information retrieval, p.42-49, August 15-19, 1999, Berkeley, California, United States [doi>10.1145/312624.312647]
• 12: Yiming Yang , Jan O. Pedersen, A Comparative Study on Feature Selection in Text Categorization, Proceedings of the Fourteenth International Conference on Machine Learning, p.412-420, July 08-12, 1997

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.5 PATTERN RECOGNITION
      I.5.2 Design Methodology
          Subjects: Feature evaluation and selection

Additional Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.3 Information Search and Retrieval

  I. Computing Methodologies
  I.5 PATTERN RECOGNITION
      I.5.2 Design Methodology
          Subjects: Classifier design and evaluation