Research Opportunity of Insider Threat Detection based on Machine Learning Methods

Insider threats have been a known threat since a long time ago in the information technology field and many researchers tried to create novel methods to solve this threat. The purpose of this paper is to find research opportunities for insider threat detection. This was done by finding and reviewing papers related to insider threat detection. The papers reviewed were only the ones that utilized machine learning algorithms because they were the most common method used by researchers to detect malicious insiders. A systematic literature review by Kitchenham, which consisted of planning, selection, extraction, and execution, was employed as the methodology. The detection method was classified into three categories: combination, selection, and singular focus. Each category discussed and recommended a research direction to create a potentially better solution for insider threat problems.