Bullet behavior abnormity detection method, storage medium, electronic equipment and system

Abstract

The invention discloses a bullet screen behavior abnormity detection method, a storage medium, electronic equipment and a system, and relates to the field of live broadcast risk control. The method comprises the steps: carrying out the statistics of all bullet screen behavior data of a live broadcast watching user account through taking a day as a unit; calculating a chi-square value ch2 accordingto each statistic behavior data; and if the calculated chi-square value is greater than the set threshold, pausing the use of the current user account, and if the calculated chi-square value is not greater than the set threshold, not performing processing. The behavior data comprises the number of sent bullet screens and the number of sent bullet screen rooms; and counting the bullet screen sending number of the user account every day and the bullet screen room sending number of the user account every day by taking the day as a unit. The abnormal account number can be accurately judged, the judgment mode is simple, and the cost is low.

Description

technical field [0001] The invention relates to the field of live broadcast risk control, and in particular to a method for detecting abnormal behavior of bullet screens, a storage medium, electronic equipment and a system. Background technique [0002] With the development of Internet technology, the functions of smart mobile devices are becoming more and more powerful. At the same time, due to the variety of live broadcast content, it is more and more popular among young people. More young people like to watch live broadcasts to enrich their spare time. Users log in to the live broadcast platform with their accounts to watch the live broadcast. [0003] However, some illegal personnel often steal user accounts and do things that violate the regulations of the live broadcast platform. These users whose accounts have been stolen are called users with abnormal accounts. For live broadcast platforms, it is necessary to ensure the viewing experience of normal users Therefore, ...

AI Technical Summary

Problems solved by technology

[0003] However, some illegal personnel often steal user accounts and do things that violate the regulations of the live broadcast platform. These users whose accounts have been stolen are called users with abnormal accounts. For live broadcast platforms, it is necessary to ensure the viewing experience of normal users Therefore, it is necessary to identify and deal with abnormal users in a timely manner. The common detection method for abnormal users is to judge whether the account has been stolen based on the user’s common device for watching live broadcasts or the usual login location. However, the accuracy of this detection method is relatively large. Large, because normal users also change devices or go to another address to watch live broadcasts, and the cost of determining common devices or common login locations is relatively high

Images