A Hybrid Convolutional Neural Network and Support Vector Machine-Based Credit Card Fraud Detection Model

Table 2

Performance analysis (%): , gamma = 0.9.

Proposed methods	Accuracy	Precision	Recall	F1-score	AUC
CNN-SVM (kernel = rbf)	91.08	90.50	90.34	90.41	91.05
CNN-SVM (kernel = linear)	90.62	94.42	87.30	90.91	90.73
CNN-SVM (kernel = sigmoid)	78.16	62.82	79.01	70.00	73.80
CNN-SVM (kernel = polynomial)	73.56	67.44	76.32	73.49	71.61