Interesting paper

[dhananjay]

The following paper is an interesting/unique application of Bayesian Networks:

Elise T. Axelrad, Paul J. Sticha, Oliver Brdiczka, and Jianqiang Shen, 2013. A Bayesian Network Model for Predicting Insider Threats. IEEE Symposium on Security and Privacy Workshops, pages 82-89.

The authors develop a BN model to predict insider threats based on several factors. These include an individual's personal characteristics, personal circumstances and professional circumstances.

It will be good to know thoughts about the model presented and also how the model is "improved" by using structural equation modelling to identify changes in the causal structure. Would structure learning methods have been more effective with the same data?

[steven.mascaro]

Those interested in the paper can find it here:

http://www.ieee-security.org/TC/SPW2013 ... 17a082.pdf