Network Security and Information Literacy Training Based on Big Data Clustering Algorithm

Abstract

Network security as a network technology platform application of the main research topic, in the complex and changeable network environment, in order to maintain network security anomaly detection is one of the main directions of practical discussion. Network security detection should first define the behavior mode of the network under the normal state, and design the model quality in advance. Assuming that the network user behavior exceeds the expected set model threshold in the network application, it is identified as the network intrusion behavior. At this point, both the design behavior model and the model threshold are the core content of network security detection, which requires rapid screening of normal user behavior patterns in a large number of network data, and the use of clustering algorithm to detect network security, so as to strengthen the information literacy of network users. This paper mainly uses ELM algorithm of PReLU activation function and improved K-means algorithm to design a multi-pole hybrid intrusion detection method. By using NSL-KDD data set to test and analyze the algorithm, it is found that compared with the traditional BP neural network algorithm, SVM algorithm and ELM algorithm, the multi-pole hybrid intrusion detection method has higher efficiency and higher accuracy, and has a strong advantage in intrusion detection judgment in network security.