Applied Marketing Science, Inc., was invited by Charles River Analytics, Inc. of Cambridge, MA to assist in the development of software that enables analysts in the Intelligence Community to forecast the accuracy and robustness of different insider threat detection systems. Insider threats are individuals within an organization enacting harmful behaviors—behaviors caused by espionage, sabotage, or even ignorance. When designing these threat detection systems, it is important to account for the uncertainty of the environment and organization in which they are placed.
The team worked under the Probabilistic Relational Inference Modeling for Enterprises (PRIME) effort to develop prediction and sensitivity analysis tools and algorithms. The effort resulted in probabilistic models of enterprise-deployed machine learning systems (such as insider threat detection systems). These models can be used to evaluate, forecast, and understand the performance of a new or existing machine learning system within an enterprise, offering an effective solution at a fraction of the cost and with greater assurance of model effectiveness than a hand-designed solution.
The project was part of the Intelligence Advanced Research Projects Activity (IARPA)’s Scientific Advances to Continuous Insider Threat Detection (SCITE) program. Read the full press release from Charles River Analytics, Inc. here.
To learn more about Machine Learning, watch our webinar on demand.
Tags: Big Data