"A Multivariate Bayesian Approach to Modeling Vulnerability Discovery in the Software Security Lifecycle," Ph.D. dissertation, The George Washington University, 2018
Dissertation presentation is available here and here
"Bayesian-Model Averaging Using MCMCBayes for Web-Browser Vulnerability Discovery," Reliability Engineering & System Safety, Volume 183, 2019, Pages 341-359
"Multivariate Models Using MCMCBayes for Web-Browser Vulnerability Discovery," Reliability Engineering & System Safety, Volume 176, 2018, Pages 52-61
MCMCBayes framework:
MCMCBAYES is a readily extensible, open-source, object-oriented
software library supporting Markov chain Monte Carlo (MCMC) based
Bayesian analyses using MATLAB. It was initially written to assist in
the evaluation of several popular fault counting models applied to
modeling software vulnerability discovery counts (Vulnerability
Discovery Models, or VDMs).
"Predicting
Post-Release Software Security Fault Discovery Using Expert Judgment
Elicitation and Bayesian Computed Effort-Dependent Models," International Doctoral Symposium on Empirical Software Engineering, October 9th, 2013, Baltimore, MD, USA