Personalized Classification for Keyword-Based Category Profiles
Personalized classification refers to allowing users to define their own categories and automating the assignment of documents to these categories. In this paper, we examine the use of keywords to define personalized categories and propose the use of Support Vector Machine (SVM) to perform personalized classification. Two scenarios have been investigated. The first assumes that the personalized categories are defined in a flat category space. The second assumes that each personalized category is defined within a pre-defined general category that provides a more specific context for the personalized category. The training documents for personalized categories are obtained from a training document pool using a search engine and a set of keywords. Our experiments have delivered better classification results using the second scenario. We also conclude that the number of keywords used can be very small and increasing them does not always lead to better classification performance.
Unable to display preview. Download preview PDF.
- 1.T. Ault and Y. Yang. kNN at TREC-9. In Proc. of the 9th Text REtrieval Conference (TREC-9), Gaithersburg, Maryland, 2000.Google Scholar
- 2.S. T. Dumais and H. Chen. Hierarchical classification of Web content. In Proc. of the 23rd ACM Int. Conf. on Research and Development in Information Retrieval (SIGIR), pages 256–263, Athens, GR, 2000.Google Scholar
- 3.S. T. Dumais, J. Platt, D. Heckerman, and M. Sahami. Inductive learning algorithms and representations for text categorization. In Proc. of the 7th Int. Conf. on Information and Knowledge Management, pages 148–155, 1998.Google Scholar
- 4.T. Joachims. SVM light, An implementation of Support Vector Machines (SVMs) in C. http://svmlight.joachims.org/.
- 5.T. Joachims. Text categorization with support vector machines: learning with many relevant features. In Proc. of the 10th European Conf. on Machine Learning, pages 137–142, Chemnitz, DE, 1998.Google Scholar
- 6.D. Koller and M. Sahami. Hierarchically classifying documents using very few words. In Proc. of the 14th Int. Conf. on Machine Learning, pages 170–178, Nashville, US, 1997.Google Scholar
- 7.K.-S. Lee, J.-H. Oh, J. Huang, J.-H. Kim, and K.-S. Choi. TREC-9 experiments at KAIST: QA, CLIR and batch filtering. In Proc. of the 9th Text REtrieval Conference (TREC-9), Gaithersburg, Maryland, 2000.Google Scholar
- 8.A. K. McCallum. BOW: A toolkit for statistical language modeling, text retrieval, classification and clustering. http://www.cs.cmu.edu/mccallum/bow, 1996.
- 9.D. Mladenic. Feature subset selection in text-learning. In Proc. of the 10th European Conf. on Machine Learning, pages 95–100, 1998.Google Scholar
- 12.S. Robertson and D. A. Hull. The TREC-9 filtering track final report. In Proc. of the 9th Text REtrieval Conference (TREC-9), Gaithersburg, Maryland, 2000.Google Scholar
- 14.TREC. Text REtrieval Conference. http://trec.nist.gov/.
- 16.Y. Yang and X. Liu. A re-examination of text categorization methods. In Proc. of the 22nd ACM Int. Conf. on Research and Development in Information Retrieval, pages 42–49, Berkeley, USA, Aug 1999.Google Scholar
- 17.Y. Zhang and J. Callan. YFilter at TREC-9. In Proc. of the 9th Text REtrieval Conference (TREC-9), Gaithersburg, Maryland, 2000.Google Scholar