Debt Detection in Social Security by Adaptive Sequence Classification
Debt detection is important for improving payment accuracy in social security. Since debt detection from customer transaction data can be generally modelled as a fraud detection problem, a straightforward solution is to extract features from transaction sequences and build a sequence classifier for debts. For long-running debt detections, the patterns in the transaction sequences may exhibit variation from time to time, which makes it imperative to adapt classification to the pattern variation. In this paper, we present a novel adaptive sequence classification framework for debt detection in a social security application. The central technique is to catch up with the pattern variation by boosting discriminative patterns and depressing less discriminative ones according to the latest sequence data.
Keywordssequence classification adaptive sequence classification boosting discriminative patterns
Unable to display preview. Download preview PDF.
- 1.Centrelink. Centrelink annual report 2004-2005. Technical report, Centrelink, Australia (2005)Google Scholar
- 2.Lesh, N., Zaki, M.J., Ogihara, M.: Mining Features for Sequence Classification. In: Proc.of the fifth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Diego, California, USA, August 1999, pp. 342–346 (1999)Google Scholar
- 3.Tseng, V.S.M., Lee, C.-H.: CBS: A new classification method by using sequential patterns. In: Proc. of SIAM International Conference on Data Mining (SDM 2005), pp. 596–600 (2005)Google Scholar
- 5.Cheng, H., Yan, X., Han, J., Hsu, C.-W.: Discriminative Frequent Pattern Analysis for Effective Classification. In: Proc. of IEEE International Conference on Data Engineering (ICDE 2007), April 2007, pp. 716–725 (2007)Google Scholar
- 7.Bonchi, F., Giannotti, F., Mainetto, G., Pedreschi, D.: A classification-based methodology for planning audit strategies in fraud detection. In: Proc. of the 5th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Diego, CA, USA, August 1999, pp. 175–184 (1999)Google Scholar
- 8.Rosset, S., Murad, U., Neumann, E., Idan, Y., Pinkas, G.: Discovery of fraud rules for telecommunications - challenges and solutions. In: Proc. of the 5th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Diego, CA, USA, August 1999, pp. 409–413 (1999)Google Scholar
- 10.Xu, J., Sung, A.H., Liu, Q.: Tree based behavior monitoring for adaptive fraud detection. In: Proc. of the 18th International Conference on Pattern Recognition, Washington, DC, USA, pp. 1208–1211 (2006)Google Scholar
- 13.Zhang, H., Zhao, Y., Cao, L., Zhang, C., Bohlscheid, H.: Customer activity sequence classification for debt prevention in social security. Accepted by Journal of Computer Science and Technology (2009)Google Scholar