Implementation and comparison of machine learning classifiers for information security risk analysis of a human resources department

The aim of this study is threefold. First, a qualitative information security risk survey is implemented in human resources department of a logistics company. Second, a machine learning risk classification and prediction model with proper data set is established from the results obtained in this survey. Third, several classifier algorithms are tested where their training and test performances are compared using error rates, ROC curves, Kappa statistics and F-measures. The results show that some classifier algorithms can be used to estimate specific human based information security risks within acceptable error rates.