PM-IUBC: A P2P and MongoDB based Intranet User Behavior Control System

Abstract

It is important for a government unit or a company to monitor the intranet security and staffs' activities during the working time. Collecting and processing massive user behaviors is difficult in the distributed intranet with thousands of hosts. In view of this, this paper proposes a P2P and MongoDB based Intranet User Behavior Control system (PM-IUBC) which consists of the user terminal, the node sliding server overlay, the database cluster and the monitor center. The user behavior control is implemented on the filter driver embedded in the user terminal, and a slice node server overlay is constructed on Pastry. A uniform light weight communication framework is designed to simplify network programming, and a MongoDB with auto shard based log database is configured to store massive user behavior logs. The user behaviors about operating software applications, removable storage devices and files can be monitored and controlled. Experiment results show that PM-IUBC can control the user behaviors and has good concurrency and scalability. 

Role Recognition Method Based on Network Flow Characteristics

“…At present, there are some studies on intranet security models or methods, such as Yu et al [5] grouping intranet user behavior using improved regular expression rules. AsSadhan et al [6] proposed a long-range Dependence (LRD) behavior analysis method based on time series analysis, and Dong et al [7] established a method. An internal network user behavior control system helps managers monitor employees' behavior.…”