Method and device for generating risk warning information and electronic equipment

Abstract

The invention relates to a method and device for generating risk warning information based on user behavior data, electronic equipment and a computer readable medium. The method comprises the steps ofobtaining behavior data of a user in a point pre-burying mode, wherein the behavior data comprises a preset behavior and corresponding time; inputting the behavior data into a word vector model to generate a user word vector; inputting the user word vector into a user risk model to generate a user risk score, the user risk model being generated through bidirectional recurrent neural network training; and when the user risk score is smaller than a threshold, generating risk warning information. According to the method and device for generating the risk warning information based on the user behavior data, the electronic equipment and the computer readable medium, historical information of a user account is not depended on, the future risk behavior of the user can be recognized only according to the user behavior data, then early warning is conducted in advance, and enterprise resource losses are reduced.

Description

technical field [0001] The present disclosure relates to the field of computer information processing, in particular, to a method, device, electronic device and computer-readable medium for generating risk warning information based on user behavior data. Background technique [0002] Financial risk prevention refers to the behavior of financial market entities to use certain methods to prevent or avoid risks in compliance with regulations on the basis of relevant analysis to achieve expected goals. In the current environment, with the increase of personal credit demand, more and more companies that provide financial services to individual users have emerged. For these financial service companies, it is important to prevent personal financial risks of users in advance. A hot technology field. [0003] In the actual risk control analysis, in the past, user risk analysis was often carried out through manual credit review. It can be judged whether to provide financial services ...

AI Technical Summary

Problems solved by technology

[0003] In the actual risk control analysis, in the past, user risk analysis was often carried out through manual credit review. It can be judged whether to provide financial services to such people based on basic information such as the user's work, income source, and handling channels, combined with business review standards. , this method has problems such as inconsistent manual standards, low review efficiency, and untrue collected information

At present, the commonly used method is to use the system credit review method to retrieve the customer's historical credit records after obtaining the customer's authorization, and judge whether to approve financial services for such people according to the resource occupation performance of the customer's existing account. This method solves all the problems of manual credit review, but it cannot correctly evaluate customers without historical accounts and creates new problems of lag

[0004] With the popularity of financial service platforms, more and more fraudulent behaviors have begun to appear. Financial service platforms are facing many fraud risks, including specific categories such as borrowing new ones, intermediaries, and intermediary agencies. After such people pretend to be users, they often The user information will be perfectly disguised, and it is difficult to identify these people only from the user information. This kind of fraudulent user will cause financial institutions to cause huge bad debt losses after subsequent lending

Images