Malicious user identification method and apparatus

Abstract

The invention provides a malicious user identification method and apparatus. The method comprises the following steps: obtaining bullet screen sending behavior features of a plurality of users; calculating a similarity value of each two users in the plurality of users according to the bullet screen sending behavior features; setting a suspect degree value of a malicious user as a first initial suspect value, and setting the suspect degree value of a to-be-identified user as a second initial suspect value; for each user, performing iterative computation on the suspect degree value of the user through a probabilistic graphical model according to the current suspect degree value of the user and the similarity value of the user and the other users; and after multiple iterative computations, using the to-be-identified user with the suspect degree value greater than a suspect threshold as the malicious user. Therefore, according to the association of the to-be-identified user and the determined malicious user on the bullet screen sending behavior features, the malicious users with unobvious malicious behavior features can be effectively screened out.

Description

technical field [0001] The present application relates to the technical field of big data processing, in particular, to a malicious user identification method and device. Background technique [0002] The live broadcast platform is an Internet social platform that provides broadcasters with a channel to display live video and interact with other users online. There are often some malicious users seeking illegitimate interests through improper means on the live broadcast platform. For example, the live video platform reflects the influence and attention of the anchor through the attention of users to the anchor. Malicious users may gain benefits by creating a large number of false attention, which will destroy the ecology of the live broadcast platform and damage the interests of other normal users. In view of this situation, malicious users need to be identified to block or display the behavior of these malicious users on the live broadcast platform. When identifying malic...

AI Technical Summary

Problems solved by technology

When identifying malicious users in the prior art, in order to prevent normal users from being identified as malicious users, stricter rules are often adopted, and users with obvious malicious behavior characteristics are regarded as malicious users, but this will miss some malicious behavior characteristics. malicious user

Images