Fraud Detection

Reference	Title	Data source (open-sourced?)	Model Type	Evaluation Metric(s)	Time Span	Primary Research Problem	Venue
Zheng et al. (2020)	Federated Meta-Learning for Fraudulent Credit Card Detection	European Credit Card (ECC) (NOT open source)，Kaggle (open source)	ResNet-34 architecture and 8-layer CNN	Compared the proposed model with 10 state-of-the-art models and outperformed all of them.	\	Fraudulent credit card detection	IJCAI-20
Wang (2020)	The Behavioral Sign of Account Theft: Realizing Online Payment Fraud Alert	3.5 million B2C transaction records from a commercial bank (NOT open source)	XGBoost (performs best), Random Forest, Logistic regression, Deep NN(3-layer)	XGB: precision 0.97, recall 0.92, f1 score 0.95,FPR:0.0005	01/04/2017- 30/06/2017	Ex-ante fraud detection, input historical transaction sequence (17 user-features and 37 online payment features), output risk score	IJCAI-20
Wang & Wellman (2020)	Market Manipulation: An Adversarial Learning Framework for Detection and Evasion	\	\	A manipulator can easily fool an existing detector with adversarially generated manipulation streams. But our regulator can detect these streams	\	Employ an adversarial learning framework which can generate market manipulating streams	IJCAI-20
Fawcett & Provost (1997)	Adaptive Fraud Detection	Records of sell phone calls by users in the New York City area	Rule learning	Accuracy 92%, Test set cost: $5403	\	One of the earliest papers that apply data mining technology on automatic fraud detection	Data Mining and Knowledge Discovery.
Chan & Stolfo (1998)	Toward Scalable Learning with Non-uniform Class and Cost Distributions: A Case Study in Credit Card Fraud Detection	A dataset from Chase Manhattan Bank	Multi-classifier meta-learning	TP,TN,FP,FN	-/10/1995- -/09/1996	One of the earliest papers that apply machine learning on credit card fraud detection	International Conference on Knowledge Discovery and Data Mining (1998)