Cost-sensitive learning and decision making for massachusetts pip claim fraud data

Authors:
Stijn Viaene;Richard A. Derrig;Guido Dedene
Affiliations:
Department of Applied Economics, Katholieke Universiteit Leuven, Naamsestraat 69, B-3000 Leuven, Belgium and Vlerick Leuven Gent Management School, Reep 1, B-9000 Gent, Belgium;Automobile Insurers Bureau of Massachusetts & Insurance Fraud Bureau of Massachusetts, 101 Arch Street, Boston, MA;Dept. of Appl Econ., Katholieke Universiteit Leuven and Vlerick Leuven Gent Mgmt. Sch., Reep 1 and Dept. of Bus. Studies, Universiteit van Amsterdam, Roetersstraat 11, 1018 WB Amsterdam, The Nethe ...
Venue:
International Journal of Intelligent Systems
Year:
2004

Citing 0
Cited 3

Asymmetric boosting

Proceedings of the 24th international conference on Machine learning
Highlighting heterogeneous samples to support vector machines' training

Neurocomputing
Hierarchical linear support vector machine

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this article, we investigate the issue of cost-sensitive classification for a data set of Massachusetts closed personal injury protection (PIP) automobile insurance claims that were previously investigated for suspicion of fraud by domain experts and for which we obtained cost information. After a theoretical exposition on cost-sensitive learning and decision-making methods, we then apply these methods to the claims data at hand to contrast the predictive performance of the documented methods for a selection of decision tree and rule learners. We use standard logistic regression and (smoothed) naive Bayes as benchmarks. © 2004 Wiley Periodicals, Inc. Int J Int Syst 19: 1197–1215, 2004.