Risk identification method and device

Abstract

The application discloses a risk identification method comprising the following steps: acquiring text communication information related to a current service operation according to the attribute information of the current service operation; extracting feature words from the text communication information to generate a word vector set; determining the risk category of the word vector set using a classification model built according to historical text communication information; and identifying whether the current service operation is at risk or not according to the risk category of the word vector set. The application further provides a risk identification device. The method provides a new way of service operation risk identification. Through risk identification according to the text communication information related to the current service operation, a service operation which may be at risk can be identified efficiently, and a powerful reference is provided for the subsequent risk control link.

Description

technical field [0001] This application relates to the field of risk control, in particular to a risk identification method and device. Background technique [0002] With the development of Internet technology, there are more and more Internet-based business operations, and most of the user's operations are normal. However, there are also a small number of fraudulent activities such as hackers or malicious users stealing other people's accounts or swiping cards. In response to this situation, risk control has become an indispensable part of the business operation process. Through risk control, on the one hand, fraudulent behavior can be avoided through early identification of risks, and on the other hand, risky business operations can be dealt with in a timely manner to avoid losses to legitimate users or minimize losses. [0003] Risk control in the prior art is usually carried out on the basis of analyzing a large amount of various types of data, for example, collecting u...

AI Technical Summary

Problems solved by technology

[0004] From the above description, it can be seen that in order to identify a small number of individual risk operations from a large number of business operations in the traditional risk control method, it is necessary to carry out risk management for all business operations one by one based on a large amount of data of various types. Rule analysis and / or model algorithm calculation, the processing process is time-consuming and inefficient. In some network business systems with a large number of business operations, the above risk control methods have been difficult to meet the online real-time application requirements

Images