Abnormal user group detection method, device and equipment based on isolated forest

Abstract

The invention belongs to the field of abnormal data analysis, and discloses an abnormal user group detection method and device based on an isolated forest, computer equipment and a readable storage medium. The method comprises the steps of encoding acquired user behavior characteristic data; carrying out dimension reduction on the coded user behavior characteristic data to obtain to-be-processed characteristic data, randomly selecting a user behavior characteristic from the to-be-processed characteristic data, and constructing an isolated forest according to a corresponding segmentation value;calculating the path length from the root node of the isolated tree to the leaf node and the average path length; and finally, calculating an abnormal score of each piece of to-be-processed characteristic data, and taking user output corresponding to the to-be-processed characteristic data of which the abnormal score is greater than a first preset value as an abnormal user; and calculating the similarity among the abnormal users, and performing grouping processing to obtain abnormal user groups. The invention further relates to a blockchain technology. The user behavior characteristics are deployed in the blockchain in a distributed manner. By adopting the method, the technical problem of inaccurate data processing and analysis is solved.

Description

technical field [0001] The present application relates to the field of abnormal data analysis, in particular to an isolated forest-based abnormal user detection method, device, computer equipment and storage medium. Background technique [0002] At present, various Internet platforms have launched various activities to attract users, but they have also attracted "wool parties" who use various tools and false information to register in batches and obtain activity rewards at the cost of performance, that is, abnormal users with gregarious characteristics . In the prior art, there are many ways to detect such abnormal users, for example, by establishing an isolated forest to detect abnormal data points and obtain abnormal users. However, this method can only screen out abnormal users from a large number of users. Sometimes the abnormal data points obtained are scattered and it is impossible to determine whether the abnormal user is an abnormal user or a group of abnormal users...

AI Technical Summary

Problems solved by technology

[0004] Based on this, it is necessary to address the above-mentioned technical problems. This application provides a method, device, computer equipment and storage medium for detecting abnormal user groups based on isolated forests, so as to solve the problem of being unable to determine whether these abnormal points are wool party in the prior art. , leading to the omission of abnormal data detection, resulting in inaccurate technical problems in data processing and analysis

Images