Communication data anomaly detection method and device, electronic equipment and storage medium

Abstract

The invention provides a communication data anomaly detection method and device, electronic equipment and a storage medium, and the method introduces a communication frequency distribution model to determine the expected communication frequency corresponding to the actual communication frequency, and judges whether the communication data is abnormal in a mode of performing goodness-of-fit detection on the actual communication frequency through the expected communication frequency. Compared with a given keyword one by one, chat records of the user do not need to be obtained, the anomaly detection efficiency can be greatly improved, and the privacy of the user can be protected. Besides, rapid detection can be carried out when the communication data is abnormal, active discovery of the communication data abnormity is realized, and the security of the to-be-detected account can be ensured.

Description

technical field [0001] The present invention relates to the technical field of data processing, in particular to a communication data anomaly detection method, device, electronic equipment and storage medium. Background technique [0002] As the communication link between people (referred to as communication) is getting closer and closer, the communication data is increasing. In order to further ensure the personal and property safety of users, abnormal detection of communication data is very important. [0003] Communication data anomaly detection refers to the detection of data that does not conform to the expected behavior pattern in the communication data. The data that does not conform to the expected behavior pattern is abnormal data. At present, there are usually three main ways to detect communication data anomalies: 1) build a keyword library, and judge whether there is an exception in the communication data through keyword filtering, text analysis, and natural lan...

AI Technical Summary

Problems solved by technology

[0004] Among the above three methods, 1) the keyword filtering method needs to be compared with the keywords in the keyword library one by one, which is inefficient and has misjudgment. Moreover, keyword filtering involves user privacy, and it is difficult for third-party tools to Get the user's chat history

2) The user builds an account blacklist, which also needs to be compared with the blacklisted accounts one by one, which is inefficient and misjudgment exists

3) Through the way of user reporting, which relies on the user's subjective judgment, the accuracy is low

Moreover, it is impossible to actively discover abnormal data in the communication data

Images