Network connection data classification method

Abstract

The invention provides a network connection data classification device for improving the classification accuracy of network connection data. The network connection data classification device providedby the invention is used for classifying W pieces of different network connection data, and comprising: setting M categories; setting a d-dimensional solution space according to d feature attribute values; randomly generating N particles in the d-dimensional solution space; separately calculating the distance between each piece of network connection data and M pieces of center data Pdhg of each particle; dividing all network connection data into M categories according to the size of the distance between each piece of network connection data and the M pieces of center data Pdhg of each particle, and performing n times of classification according to n data points; calculating the sum of the distances between all network connection data and the corresponding center data Pdhg in each classification; judging the particle having the minimum distance sum in the N times of classification; adjusting the current positions and the current change rates of the other particles; judging whether to terminate the classification; and outputting a result.

Description

technical field [0001] The invention specifically relates to a method for classifying network connection data. Background technique [0002] With the explosive popularity of the Internet in recent years, the Internet has penetrated into people's life, entertainment and work. However, the openness and security of the Internet is a double-edged sword. While it brings convenience to people, the Internet's non-administrative, transnational, and undefended nature makes network security problems more and more prominent. Network intrusion detection is an important part of network security system. It monitors and responds to unauthorized use and abuse of network resources, and has the function of protecting information integrity and confidentiality. [0003] Generally speaking, network intrusion detection methods include anomaly intrusion detection and misuse intrusion detection methods. Misusing the intrusion detection method considers that there is a large intersection between a...

AI Technical Summary

Problems solved by technology

[0005] In order to ensure the accuracy of data mining and reduce the false alarm rate, it is necessary to construct an accurate classification of network connection data in advance, but in the process of constructing data classification, it is often easy to fall into the problem of local optimum, resulting in quite inaccurate classification

Images