Credit card fraud detection method and device based on multi-feature fusion and medium

The invention provides a credit card fraud detection method and device based on multi-feature fusion and a medium, which can effectively improve the accuracy and recall rate of fraudulent transaction prediction and greatly reduce the false positive rate of fraudulent transactions, and the method comprises the following steps: collecting cardholder information data; performing embedded representation on the basic information data, the financial operation behavior data and the credit card transaction information data of the cardholder through a feedforward neural network model, a convolutional neural network model and a Bi-LSTM model, and performing splicing to obtain a sample feature representation vector; constructing and training a generative adversarial network model; using the trained generator model to generate a simulated fraud sample, combining the simulated fraud sample to generate a new training set, constructing and training a plurality of detection base models, and combining the trained detection ba