Abnormal application detection method and device, computer device and storage medium

Abstract

The invention relates to the field of artificial intelligence classification models, and specifically relates to an abnormal application detection method and device, a computer device and a storage medium. The method comprises the steps of obtaining credit application data of an applicant; obtaining applicant information and an application feature set corresponding to the applicant information according to the credit application data; inputting the application feature set into a preset isolated forest model, obtaining an application score corresponding to the application feature set, and constructing the preset isolated forest model based on an unmarked training feature set; and judging whether an application corresponding to the credit application data belongs to an abnormal application or not according to the application score. The credit application is scored by adopting the preset isolated forest model obtained by unsupervised learning, no label is needed for training, the practicability of the abnormal application scoring system is greatly improved, variable fraud and unseen fraud can be recognized, the abnormal degree of the abnormal application scoring system is output in ascoring mode, and the abnormal degree of the abnormal application scoring system is reflected, so that clients can understand and productize the abnormal application conveniently.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to a method, device, computer equipment and storage medium for detecting abnormal applications. Background technique [0002] The online credit industry has developed rapidly in recent years, presenting a situation where a hundred schools of thought are contending and a hundred flowers are blooming. Along with the prosperous development of the industry, the black industry chain of fraud is constantly penetrating into this field, and various novel fraud modes emerge in endlessly. The healthy development of the Internet credit industry has cast a shadow. According to incomplete statistics, the annual loss due to fraud is 50 billion to 100 billion, and the risk of fraud has become the most important risk in the Internet credit industry. [0003] At present, the industry mainly uses supervised learning algorithms to detect fraudulent loan applications from users, but in mos...

AI Technical Summary

Problems solved by technology

[0003] At present, the industry mainly uses supervised learning algorithms to detect fraudulent loan applications from users, but in most cases, the data used to detect whether user behavior is fraudulent is unlabeled, and the cost of manual labeling is extremely high, and the means of fraud always changing

Images