Research On User’s Traffic Behavior Analysis Method Based On Network Traffic Monitoring And Prediction

While Internet providing convenience to user’s daily life, however, networksecurity problems appear at the same time. Existing network security technology aremainly focus on defensive response for foreign attack, while consider less for theabnormal traffic behavior of the internal network users. Therefore, a full range ofmonitoring for user’s network traffic behavior is a essential part to ensure that thesecurity of network. In this paper. We mainly do research on the analysis of networkuser’s traffic behavior.The problems of existing network user’s traffic behavior analysis system arelacking of a systematic definition of network user’s traffic behavior. The descriptiondimension to network user’s traffic behavior is too high, so it is time-consuming toanalyze single network user’s traffic behavior from massive network data. Toovercome the above shortcomings, we first research the method to establish a singlenetwork user’s traffic behavior model, and a modeling method based on associationrules is proposed for the establishment of a stable normal user’s traffic behaviorpattern. Then we establish a more complete feature set based on the characteristic ofnetwork traffic to describe network user’s behavior. in addition, a feature selectionrule based on the relative deviation distance is proposed to select the optimizedfeature set. We use the selected feature set to locate the abnormal moment and userswho produce abnormal traffic behavior. Finally, a traffic behavior analysis methodbased on prediction is proposed to predict the future features value so as to improveefficiency of the system. The experimental results show that the proposed method hasan excellent detection rate and efficiency in the analysis of abnormal user’s trafficbehavior.