Anti-fraud detection method and device, computer equipment and storage medium

Abstract

The invention relates to an anti-fraud detection method and device for an insurance claim settlement case, computer equipment and a storage medium. The method comprises the steps of obtaining a plurality of to-be-processed images, an image feature vector corresponding to each to-be-processed image and a case association parameter; determining every two target to-be-processed images which are similar to each other; generating a corresponding text feature vector according to the case association parameter of each target to-be-processed image; fusing the image feature vector and the text featurevector corresponding to each target to-be-processed image to obtain a target feature vector; and determining whether the claim settlement case is a fraudulent case or not according to the target feature vectors of every two target to-be-processed images. According to the method, the similar images are automatically screened based on the similarity between the images, fraudulent cases are judged based on the similar images, and the claim settlement case detection efficiency is improved; and whether the claim settlement case is a fraudulent case is detected by combining the case association parameter features and the image features of the similar images, so that the claim settlement case detection precision is improved.

Description

technical field [0001] The present application relates to the field of computer technology, and in particular, to an anti-fraud detection method, device, computer equipment and storage medium for insurance claims. Background technique [0002] With the continuous development of the social economy and the continuous improvement of people's living standards, the number of motor vehicles has increased rapidly. The rapid growth of motor vehicles has prompted the development of the auto insurance industry. Auto insurance (referred to as auto insurance) refers to a commercial insurance that is liable for personal injury or property damage caused by natural disasters or accidents. [0003] In the auto insurance claims business, several claims applications may be received for accidents at the same location, and there is a risk of insurance fraud. In the related art, the screening of such fraudulent behaviors is usually performed during the review of auto insurance claims, and the ...

AI Technical Summary

Problems solved by technology

However, since different cases will be randomly distributed to different reviewers for review, it is not accurate enough to judge whether the case is the same accident scene by manually reviewing pictures, and then judge whether there is a suspicion of fraud.

Images