Learned lessons in credit card fraud detection from a practitioner perspective

By Andrea Dal Pozzolo, Olivier Caelen, Yann-Aël Le Borgne, Serge Waterschoot, Gianluca Bontempi. Expert systems with applications, 2014

537 citations recorded by the source corpus.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/6d2e2a1caf5b3757ed0e8f404eabb31546d5698a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Gianluca Bontempi
2. Andrea Dal Pozzolo
3. Sergey I. Nikolenko
4. Olivier Caelen
5. A. Taha
6. Sharaf Jameel Malebary
7. Leman Akoglu
8. Cesare Alippi

Citations in the corpus

• Adaptive Machine Learning for Credit Card Fraud Detection (2015, methodology): “Frauds represent a small fraction of all the daily transactions [18].”
• An Intelligent Approach to Credit Card Fraud Detection Using an Optimized Light Gradient Boosting Machine (2020, background): “The AUC curve is an important and useful estimation of overall performance and a general measure of the accuracy of fraud investigation [29].”
• APATE: A novel approach for automated credit card transaction fraud detection using network-based extensions (2015, background): “While e-commerce is still gaining popularity, it is also the playground of fraudsters who try to misuse the transparency of online purchases and the transfer of credit card records.”
• Synthetic Data for Deep Learning (2019, methodology): “As a specific application, the authors use existing fraud detection algorithms to find anomalies in the Kaggle Credit Card Fraud Detection Dataset [136] and generate synthetic data that balances the found abnormalities.”
• Feature engineering strategies for credit card fraud detection (2016, methodology): “Differnt detection systems that are based on machine learning techiques have been successfully used for this problem, in particuar: neural networks (Maes, Tuyls, Vanschoenwinkel, & Manderick, 002), Bayesian learning (Maes et al., 2002), artificial immune sysems (Bachmayer, 2008), association rules (”
• Calibrating Probability with Undersampling for Unbalanced Classification (2015, methodology): “To avoid the problem of different misclassification costs, we use an estimation of AUC based on the MannWhitney statistic [10].”
• Credit Card Fraud Detection: A Realistic Modeling and a Novel Learning Strategy (2018, methodology): “The majority of works addressing the fraud-detection problem in credit card transactions (see [5], [23], [63]) unrealistically assume that the class of each transaction is immediately provided for training the classifier.”
• Fraud detection system: A survey (2016, methodology): “In addition, fraudster's tricks are continuously evolving and detection has to adapt to these new fraud types (Dal Pozzolo et al., 2014).”
• Learning temporal representation of transaction amount for fraudulent transaction recognition using CNN, Stacked LSTM, and CNN-LSTM (2017, methodology): “Many methods have been proposed such as: Random forest [4] [5], Decision Tree [6], K-Nearest Neighbors [7], SVM [8] [9], HMM [10], neural networks [11], Bayesian learning [11], artificial immune systems [12], association rules [13], hybrid models [14], discriminant analysis [15].”