Research Area: CyberSecurity
The application of machine learning (ML) technique in cyber-security is increasing than ever before. Starting from IP traffic classification, filtering malicious traffic for intrusion detection, ML is the one of the promising answers that can be effective against zero day threats. New research is being done by use of statistical traffic characteristics and ML techniques. This paper is a focused literature survey of machine learning and its application to cyber analytics for intrusion detection, traffic classification and applications such as email filtering. Based on the relevance and the number of citation each methods were identified and summarized. Because datasets are an important part of the ML approaches some well know datasets are also mentioned. Some recommendations are also provided on when to use a given algorithm. An evaluation of four ML algorithms has been performed on MODBUS data collected from a gas pipeline. Various attacks have been classified using the ML algorithms and finally the performance of each algorithm have been assessed.
Keywords:
Machine learning
Data mining
cyber security
Author(s) Name: Rishabh Das; Thomas H. Morris
Journal name:
Conferrence name: 2017 International Conference on Computer, Electrical & Communication Engineering (ICCECE)
Publisher name: IEEE
DOI: 10.1109/ICCECE.2017.8526232
Volume Information:
