An Interactive Visualization System for Network Security Data
Li Zheng,Gang Yu,Yuntian Zheng
DOI: https://doi.org/10.1007/s11265-023-01836-0
2023-02-01
Journal of Signal Processing Systems
Abstract:Data visualization is an important approach for data analysis, which can reveal the patterns and characteristics of complex datasets through visual processing, and provide aid for data analysts. In recent years, with the expansion of Internet users and the rich diversity of various Internet applications, the importance of network security is increasing. In the field of network security data analysis, it is a developing direction to use data visualization methods and visual analysis tools to assist manual analysis. In this paper, a visualization system for network security data is designed and implemented, which is mainly based on network flow records and security policies. The node centrality measurement and high-dimensional data visualization methods are comprehensively applied, and an interactive visualization approach is proposed. The IP topology is presented in three different view modes: static analysis, temporal analysis and exploration analysis. In addition, the high-dimensional projection map takes flow, security policy and IP address as analysis objects and selects several dimensional features as indicators. After visual presentation, the distribution of the stream set and IP set contained in the record after projection in the selected dimension space can be obtained, so that the analyst can find the points with significant abnormal eigenvalue distribution, and deduce the possible situation based on this. After testing, the system can accept the new original data record file, and quickly generate the corresponding visualization content, which can be used for the visualization analysis of network security optimization tool software, and has been used in the actual system.
computer science, information systems,engineering, electrical & electronic