Design And Implementation Of An Adaptive User And Entity Behavior Analysis System

In recent years,with the rapid development of network technology,network information security has attracted more and more attention.Cyber attacks are also increasing and the attack methods are becoming more and more complex,In particular,internal attacks,unlike external attacks,often originate from the intentional or unintentional behavior of internal users,many abnormal behaviors are often disguised in the normal behavior,with the characteristics of concealment and destructiveness.As well as causing more damage to businesses,it is also less likely to be detected.Therefore,how to detect and prevent internal threats has become increasingly important.With the rapid development of deep learning technology and the major breakthroughs made in recent years,the user behavior analysis technology has been constantly improved,providing a lot of effective detection and corresponding schemes for abnormal attacks.Including the use of the user's behavior characteristics,subjective factors,motivation to diagnose the abnormal degree of user behavior,use the user role,objective factors such as access to detect abnormal behavior of users,and the use of machine learning approach for large data mining through the result analysis of user behavior whether there is abnormal.In this paper,from the Angle of the internal network structure,the internal network of entities and their relationships were analyzed,and the portraits of behavior on the real data and traffic data and description,quantitative relationship between entities,and entity of internal network anomaly detection,finally put forward a kind of based on adaptive user and the real threat of internal network behavior analysis system for testing and discovery.