RESEARCH ON SOFTWARE USER BEHAVIOR CREDIBILITY ANALYSIS MODEL BASED ON MULTI-STRATEGY LEARNING ALGORITHM

It is necessary to evaluate the credibility of software user behavior for the timely identification and control of abnormal user behavior risks in order to ensure the security of the software system and the credibility of user behavior. The user behavior log dataset is generated by using the simulated county and city government office system, and the credibility analysis algorithm model of numerous individual learning and ensemble learning are constructed to evaluate the effect of the credibility analysis model in the study, and by comparing and analyzing the heterogeneous ensemble learning algorithm models of different combinations, the accuracy of software user behavior credibility analysis has been improved to more than 97%. This study shows that the model is able to identify untrusted users in a flexible yet accurate manner and guarantees the security of the fixed domain work system.