DaWaK 2009: Data Warehousing and Knowledge Discovery pp 428-440 | Cite as
Rule Learning with Probabilistic Smoothing
Abstract
A hierarchical classification framework is proposed for discriminating rare classes in imprecise domains, characterized by rarity (of both classes and cases), noise and low class separability. The devised framework couples the rules of a rule-based classifier with as many local probabilistic generative models. These are trained over the coverage of the corresponding rules to better catch those globally rare cases/classes that become less rare in the coverage. Two novel schemes for tightly integrating rule-based and probabilistic classification are introduced, that classify unlabeled cases by considering multiple classifier rules as well as their local probabilistic counterparts. An intensive evaluation shows that the proposed framework is competitive and often superior in accuracy w.r.t. established competitors, while overcoming them in dealing with rare classes.
Keywords
Association Rule Minority Class Decision Region Default Rule Training CasePreview
Unable to display preview. Download preview PDF.
References
- 1.Agrawal, R., Srikant, R.: Fast algorithms for mining association rules. In: Proc. of Int. Conf. on Very Large Data Bases, pp. 487–499 (1994)Google Scholar
- 2.Antonie, M.-L., Zaïane, O.R.: Text document categorization by term association. In: Proc. on IEEE Int. Conf. on Data Mining, pp. 19–26 (2002)Google Scholar
- 3.Arunasalam, B., Chawla, S.: CCCS: A top-down association classifier for imbalanced class distribution. In: Proc. of ACM SIGKDD KDD, pp. 517–522 (2006)Google Scholar
- 4.Asuncion, A., Newman, D.J.: UCI machine learning repository (2007)Google Scholar
- 5.Bishop, C.M.: Pattern Recognition and Machine Learning. Springer, Heidelberg (2006)MATHGoogle Scholar
- 6.Cesario, E., Folino, F., Locane, A., Manco, G., Ortale, R.: Boosting text segmentation via progressive classification. Knowledge and Information Systems 15(3), 285–320 (2008)CrossRefGoogle Scholar
- 7.Coenen, F.: LUCS KDD implementations of CBA and CMAR (2004)Google Scholar
- 8.Cohen, W.W.: Fast effective rule induction. In: Proc. of Int. Conf. on Machine Learning, pp. 115–123 (1995)Google Scholar
- 9.Fawcett, T.: An introduction to ROC analysis. Pattern Recognition Letters 27(8), 861–874 (2006)MathSciNetCrossRefGoogle Scholar
- 10.Frank, E., Witten, I.H.: Generating accurate rule sets without global optimization. In: Proc. of Int. Conf. on Machine Learning, pp. 144–151 (1998)Google Scholar
- 11.Han, J., Yin, Y.: Mining frequent patterns without candidate generation. In: Proc. of ACM SIGMOD Int. Conf. on Management of data, pp. 1–12 (2000)Google Scholar
- 12.Holte, R.C., Acker, L., Porter, B.: Concept learning and the problem of small disjuncts. In: Proc. of Int. Joint Conf. on Artificial Intelligence, pp. 813–818 (1989)Google Scholar
- 13.Li, W., Han, J., Pei, J.: CMAR: Accurate and efficient classification based on multiple class-association rules. In: Proc. of IEEE Int. Conf. on Data Mining, pp. 369–376 (2001)Google Scholar
- 14.Liu, B., Hsu, W., Ma, Y.: Integrating classification and association rule mining. In: Proc. of ACM SIGKDD Int. Conf. on Kwnoledge Discovery and Data Mining, pp. 80–86 (1998)Google Scholar
- 15.Liu, B., Ma, Y., Wong, C.K.: Improving an association rule based classifier. In: Proc. of Principles of Data Mining and Knowledge Discovery, pp. 504–509 (2000)Google Scholar
- 16.Thabtah, F.: A review of associative classification mining. The Knowledge Engineering Review 22(1), 37–65 (2007)CrossRefGoogle Scholar
- 17.Webb, G., Boughton, J., Wang, Z.: Not so naive bayes: Aggregating one-dependence estimators. Machine Learning 58(1), 5–24 (2005)CrossRefMATHGoogle Scholar
- 18.Weiss, G.M.: Mining with rarity: A unifying framework. ACM SIGKDD Explorations 6(1), 7–19 (2004)CrossRefGoogle Scholar
- 19.Xin, X., Han, J.: CPAR: Classification based on predictive association rules. In: Proc. of SIAM Int. Conf. on Data Mining, pp. 331–335 (2003)Google Scholar