Measuring the Quality Information of Sources of Cybersecurity by Multi-Criteria Decision Making Techniques

Abstract

Companies and public institutions handle a large amount of data that could affect the quality of the information. This can produce that any decision-making process based on the data is not good enough. In this article, we present a decision support software to assess the quality of the data of cybersecurity data sources integrating multi-criteria decision-making techniques. These methods let us rank the sources in terms of the quality of the information that they provide. First of them is the well-known Weighted Sum Model. Since the evaluation of the cybersecurity data sources has to be flexible to the institution’s policies, we have adapted the application of the Analytic Hierarchy Process method. Also, we have included the computation of Goodman and Kruskal’s coefficient to measure the concordance between the obtained rankings. We have evaluated the data collected by the cyber data sources through the usual records received in a cybersecurity incident response team. The study was carried out over a real dataset containing 25, 297, 210 cybersecurity event records, 27 sources, and 55 types of cybersecurity events. From the results, we have performed the diagnosis identifying those data aspects that can be improved.