Advancing Anomaly Detection in Tabular Data: A Case-Study on Credit Card Fraud Identification

Recent advances in the field of Machine Learning has enabled banks to rely on this class of algorithms to build or augment their detection systems. Nevertheless, applying machine learning methods to identify frauds still remains challenging due to (i) the inherent imbalance in the available datasets and (ii) the possibility of distribution shift. Weakly-supervised anomaly detection (AD) methods appear as a possible solution as they should be robust to both challenges. In this work, we propose two novel weakly-supervised AD methods targeted for tabular data. We then testDéfinition courte Lorem ipsum AD methods on a private online credit card payment dataset and compare their performance to Gradient Boosted Decision Trees (GBDT). 

We observe a significant performance gap between GBDT and AD methods, in favor of GBDT. Our experiments supports the idea that although promising, weakly-supervised AD method need further improvements to compete with GBDT for the task of fraud detection.

Jury

• Alain CELISSE, rapporteur.
• Marius KLOFT, rapporteur.
• Mazen ALAMIR, examinateur.
• Louise TRAVÉ-MASSUYÈS, examinatrice.
• Gaël VAROQUAUX, examinateur.