Abnormal account detection model training method and abnormal account detection method

Abstract

The invention provides an abnormal account detection model training method and an abnormal account detection method, and the abnormal account detection model training method comprises the steps: receiving a training sample which comprises target accounts and task routes; inputting each target account and each task route into an abnormal account detection model, and clustering each task route according to a first similarity threshold to obtain a task route cluster; counting a first number of task routes corresponding to the abnormal account in the abnormal task route cluster; identifying account states of unlabeled accounts in the abnormal task route cluster, and counting a second number of unlabeled accounts with abnormal account states and a third number of unlabeled accounts with normal account states in the abnormal task route cluster; and adjusting the first similarity threshold value and the abnormal threshold value according to the first number, the second number and the third number, returning to execute the clustering step until a training stop condition is met, and storing the cluster center of the abnormal task route clustering cluster. The efficiency of detecting abnormal accounts can be improved.

Description

technical field [0001] The present application relates to the field of computer technology, and in particular to an abnormal account detection model training method and an abnormal account detection method. Background technique [0002] With the rapid development of computer technology, various games emerge in an endless stream. In the field of games, massively multiplayer online role-playing games are the most popular, and many studios have emerged for such games, such as Dajin Studio and Power Leveling Studio. In these studios, high-end game players or enthusiasts use a large number of high-end configuration computers to run plug-in scripts to play games, and collect real money to help players earn game currency and leveling. These two businesses will be upgraded through target accounts that use a large number of plug-in scripts. Process tasks. This behavior will directly have a negative impact on other normal players, disrupting the game environment and economic balance...

AI Technical Summary

Problems solved by technology

However, the labor cost of the above method is high, and due to the change of the plug-in script, the judgment using numerical features will often fail, resulting in low efficiency and low accuracy in detecting abnormal accounts

Images