Impact of False Positives and False Negatives on Security Risks in Transactions Under Threat

This paper presents a theoretical model, algorithms, and quantitative assessment of the impact of false positives and false negatives on security risks during transaction processing. It is based on analysis of the effect of the parameters of the optimal strategy for planning countermeasures in Partially Observable Markov Decision Processes. The paper reveals the dependency between the false positives and false negatives of the analytics and their impact on the efficacy of the algorithms for detection. Such an analysis is an important component of cybersecurity frameworks, planning countermeasures for mitigating the risks from security threats, but it can be applied to many other business processes which experience device malfunctions or human errors.