Research on Baseline Technology of Industrial Control Network Security based on Semi-supervised Learning

Abstract

With the rapid development of industrial control network, performance management and risk prevention based on network traffic data, especially abnormal traffic detection, have gradually attracted people's attention. However, the traditional flow detection method based on fixed baseline cannot adapt to the growing data and increasingly complex data types. It leads to inaccurate test results and false alarms, and also consumes a lot of manpower and resources. In this paper, a semisupervised learning method is proposed to realize the self-construction of baseline and the automatic detection of abnormal index data.