Insider threat detection using principal component analysis and self-organising map

An insider threat can take on many aspects. Some employees abuse their positions of trust by disrupting normal operations, while others export valuable or confidential data which can damage the employer's marketing position and reputation. In addition, some just lose their credentials which are then abused in their name. In this paper, we use Principal Component Analysis (PCA) in conjunction with Self-Organising Map (SOM) for insider threat detection within an organisation. The results show that using PCA before SOM increases the clustering accuracy.