Tree-based Cost-Sensitive Methods for Fraud Detection in Imbalanced Data

Guillaume Metzler (1) , Xavier Badiche (2) , Brahim Belkasmi (2) , Elisa Fromont (3) , Amaury Habrard (1) , Marc Sebban (1)

1 LHC - Laboratoire Hubert Curien
2 Blitz Business Service
3 LACODAM - Large Scale Collaborative Data Mining

Bank fraud detection is a difficult classification problem where the number of frauds is much smaller than the number of genuine transactions. In this paper, we present cost sensitive tree-based learning strategies applied in this context of highly imbalanced data. We first propose a cost sensitive splitting criterion for decision trees that takes into account the cost of each transaction and we extend it with a decision rule for classification with tree ensembles. We then propose a new cost-sensitive loss for gradient boosting. Both methods have been shown to be particularly relevant in the context of imbalanced data. Experiments on a proprietary dataset of bank fraud detection in retail transactions show that our cost sensitive algorithms allow to increase the retailer's benefits by 1,43% compared to non cost-sensitive ones and that the gradient boosting approach outperforms all its competitors.

Domaines: Intelligence artificielle [cs.AI]; Apprentissage [cs.LG]

Guillaume Metzler, Xavier Badiche, Brahim Belkasmi, Elisa Fromont, Amaury Habrard, et al.. Tree-based Cost-Sensitive Methods for Fraud Detection in Imbalanced Data. IDA 2018 - 17th International Symposium on Intelligent Data Analysis, Oct 2018, ‘s-Hertogenbosch, Netherlands. pp.213-224, ⟨10.1007/978-3-030-01768-2_18⟩. ⟨hal-01895967⟩