Mining User Behavior Rules Based On Network Packet

Network management needs to monitor 、 test、 configurate、 analyse、 evaluate and control network resources, and report and process the network faults in time, and coordinate and maintain efficient operation of the network system. Therefore, you need to get the network traffic、 analysis it and give a optimization scheme. Data mining can find out some hidden information from a large amount of data through different kinds of algorithms. As the explosive growth of network information, as well as the ability of the computer processing and storaging capacity increasing, we have basic conditions to apply data mining technology to deal with the network traffic information.In fact, dealing with huge amounts of information in the network, has been used in many fields. From a security point of view to consider, all kinds of firewalls, antivirus softwares and intrusion detection systems are developing continuously. From a business perspective, the electronic commerce vigorously developed in recent years.Using the Internet users’ behaviors in e-commerce network and personal information provided by users to find potential customer group technology is also developing continuously. From the point of view of network management, network management should do their best to make the network security、stable、efficiency and healthy. Therefore, the effective analysis of network traffic is the key to make full use of network massive amounts of information.Network traffic is essentially the sending and receiving data packets through the network to accomplish the network interaction. All of the information of the network users is in the network packets. As information amount of the network packets is very large, it is suitable for the data processing object as data mining. Therefore, you can capture network packets, analyse the protocol parsing of network packets, use data mining technology in the network packet, and then find the hidden knowle dge in them.With the development of science and technology, the Internet has been made up of desktop computers, laptops, tablet computers, mobile phones and other intelligent equipment. Each device has a unique MAC address. As for the Internet,they have no difference in network communication and data transmission from the traditional way of computer Internet. For this reason, we can also capture network packets from these intelligent equipment.This paper mainly introduces a kind of tools based on zero copy called netsniff-ng, suitable for passive packet capture in high-speed network, capturing network packets, and analysing the network packet protocol parsing,especially in application layer. Finally using Association analysis in data mining to find out user behavior rules.