Intelligent Detection Method of Computer Network Intrusion based on Big Data Clustering Algorithm

Abstract

N etwork technology is rapidly developing and the Internet has penetrated into every step of people's daily production life. As the importance of the Internet continues to strengthen, security issues are becoming increasingly acute. And the network security problem in the context of big data (BD) presents the characteristics of new mode, large scale and high concealment. Therefore, the research of IDM based on BD features has received wide attention in the field of network security and is applied in various fields. In this paper, we propose an intelligent detection method for computer network intrusion based on BD clustering algorithm for the lack of relevance, timeliness and targeting of current computer network intrusion event detection technology. Firstly, the network intrusion events are classified, and the contents contained in the target files are obtained by clustering algorithm clustering, then the files are clustered and analyzed using neural network, and the intrusion events are classified according to the classification results after clustering. The ID results implemented based on the clustering algorithm can be compared and analyzed with traditional methods. Firstly, the data is pre-processed in the data mining module and then the data set is constructed for clustering training; then the data is clustered after training set by cross-validating and calculating the classification results. After training to get the sample set again to cluster the data training set samples using Number Stata 2 training set data using randomly generated sample dataset; finally get the model dataset using the algorithm proposed in this paper to classify the target documents. In the experiment detection results show that: the intrusion strategy obtained after clustering analysis is significantly more accurate compared with the ID results; detection accuracy can reach 93.3%; has the advantages of good detection effect and detection speed.