Subjective logic application research-Rough set belief rule model and trust evaluation model of the intrusion detection system

摘要

This paper proposed a rough set belief rule model and a trust evaluation model of the intrusion detection system for subjective logic application research. The rough set belief rule model builds the belief rule expectation for classification and recognition. It is not only considered the effects of single condition attributes to decision, but also the multi-attributes. The example analysis and comparing with the R-0.5(X) model indicates the validity and advantage of this model. In evaluation model, aiming at the intrusion detection system of two types of errors (false positive and false negative), the trust evaluation of the whole intrusion detection system is given. It is not only from the macro, but also for each rule. According to the trust of each rule, the corresponding rules are selected again for the lower trust rules. So, the discern ability of intrusion detection system would be improved. The simulation experiment results show that the model can effectively evaluate the trust of the intrusion detection system. (C) 2015 Elsevier GmbH. All rights reserved.