Using K-means Clustering Algorithm to Identify Network Attack Patterns in Big Data Environment

With the rapid development of network technology, network security threats have become increasingly complex and diverse. Traditional network attack detection based on rules or statistical methods can no longer meet the high efficiency and real-time requirements in the big data environment. This paper proposes a network attack pattern recognition method based on K-means clustering algorithm to detect and analyze abnormal behaviors in large-scale network traffic data. Through improvements to the K-means algorithm, including K-means++ initialization, dynamic cluster number determination, and combination with a distributed computing framework, the clustering effect and operating efficiency of the algorithm have been significantly improved. In addition, this paper performs feature extraction and dimensionality reduction on network traffic data, uses t-SNE to visualize the clustering results, and intuitively displays the distribution differences between normal traffic and abnormal traffic. Experimental results show that the improved K-means algorithm is better than the traditional clustering algorithm in terms of accuracy and other indicators, and has good scalability in the big data environment. This article provides an efficient and scalable technical solution for network attack detection, and provides strong support for the intelligent development of network security protection.