Analysis Of Enterprise Network User Behavior

With the rapid development of Chinese LAN (Local Address Network), more and more enterprises are building their own LAN, called enterprise intranet, to complete the progress of enterprise informatization. During the ever-growing and ever-complicated enterprise network, the problems of how to manage the network, and how to improve the staffs work and enterprise operation efficiency are more and more crucial for enterprise network management and business analysis of internet applications.In this thesis, we firstly introduce the conceptions of network user behavior and network traffic monitoring. And then we mainly present the integrated management system of enterprise intranet, which is an integrated system monitoring and controlling the intranet, including both the hardware device of network traffic monitoring and powerful background support software system and management platform, for the purpose of improving the efficiency of enterprise management and reducing the expense of network operation.Network application identification is the base of analysis of user behavior, for that the original data needed in analysis of user behavior must be distinguished from the level of network application. In this thesis, we have discussed the main methods of network application identification-DFI and DPI, and then we have analyzed the technology of network application identification deeply, combined with some concrete cases of network application identification, to lay the better foundation for analysis of user behavior in next step.To make better analysis of enterprise network user behavior, we can take methods from different aspects and different levels. From the aspect of network traffic, we have analyzed network traffic generated by different network applications, including global clustering analysis and average clustering analysis. From the aspect of user behavior quantity, we have analyzed the quantities of different user behaviors. Finally, we focus on the web surfing analysis to capture the users’most common behavior. We examine the web surfing classification data both on individual level and network level.12classes of web pages are taken into consideration and the data of more than10thousand of users from188enterprises over an epoch of15days are collected. For every subject under examination, the web surfing frequency of12classes forms a12-dimensional vector. We cluster all the vectors on different levels and the users and enterprises are classified respectively. To evaluate the stability of user’s behavior, we introduce a stability measure for web surfing behavior. We calculate this measure for all the users and the results are analyzed. The web surfing behavior classification and behavior stability analysis are very useful for the enterprise manager to find out inappropriate use of the enterprise network.