RISK-CONTROL MODEL TRAINING METHOD, RISK IDENTIFICATION METHOD AND APPARATUS, AND DEVICE AND MEDIUM

摘要

A risk-control model training method, a risk identification method and apparatus, and a device and a medium. The risk-control model training method comprises: labeling original video data to obtain positive and negative samples (S11); carrying out frame dividing and facial detection on the positive and negative samples to obtain training facial images (S12); grouping the training facial images according to a pre-set number to obtain at least one group of target training data (S13), wherein the target training data comprise a consecutive N frames of training facial images; dividing the target training data according to a pre-set ratio to obtain a training set and a test set (S14); inputting each group of target training data in the training set into a convolutional neural network-long-short term recurrent neural network model to carry out training to obtain an original risk-control model (S15); and using each group of target training data in the test set to test the original risk-control model to obtain a target risk-control model (S16). The risk-control model training method has the advantages of high training efficiency and high identification precision.