Insider Threat Detection with Text Libraries and Machine Learning

Kane, Patrick Ryan; Li, Bryan Z.

Student Work

Insider Threat Detection with Text Libraries and Machine Learning

Public

In an effort to detect insider threats, we researched the possibility of using the text from a user’s screen to detect anomalies in user behavior and classify certain text as sensitive. By monitoring the text on a user’s screen as well as the different interface protocols used by the computer, we are able to capture every useful action committed on a computer at any given time. This proved an effective approach because by detecting abnormal behavior, we are able to record the incident, where an auditor can then view the scenario where the behavior took place, and take the appropriate actions. The Insider Threat Detection System provides an extra layer of security for preventing unauthorized data from being released outside a company’s secure internal network.

This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.

Creator

Publisher