Machine Learning Approaches in Cybersecurity

Abstract

Machine Learning is one of the effective responses to trivial attacks, starting with the Internet Protocol traffic classification and the filtering of mistreatment traffic for intrusion detection. The latest study is being conducted out using traffic data and Machine Learning techniques. This article presents a study of the literature on machine learning and its uses in online-based data security frameworks for malware prevention, labeling, and utilization like email filtering. Each approach has been defined and synthesized based on the significance and amount of citation. Due to the importance of datasets in the Machine Learning methods, many well-known databases are often referenced. There are also some guidelines for using a particular algorithm. The MODBUS data obtained from a gas pipeline were evaluated using four Machine Learning algorithms. Several attacks were categorized by Machine Learning algorithms and each algorithm was ultimately judged for its efficiency.