Abstract
Insider threats, is one of the most challenging threats in cyberspace,
usually responsible for causing significant loss to organizations.The
topic of insider threats has long been studied and many detection
techniques were proposed to deal with insider threats. This paper
focuses on using different anomaly detection algorithms- Locality
Outlier Factor Algorithm and Isolation forest Algorithm and does a
comparative analysis between their performance. A hybrid model
incorporating advantages of both LOF Algorithm and IF Algorithm is
proposed in this paper which gives better performance than the
individual models for detecting insider threats. The hybrid model was
able to achieve whooping 99.99\% accuracy while
detecting insider threats.