A semantic framework for a better understanding, investigation and prevention of organized financial crime

Abstract

Using semantic technology for data storage and exploration is an important issue in computer science, however barely applied to forensic investigations. In this paper, a conceptual framework is proposed for the detailed modeling of structured domain knowledge in the field of organized financial crime, with a special focus on sparse information (e.g. flows of money, data and know-how, exploited vulnerabilities and attackers motivation) and the proposition of a credibility measure (to rate the reliability of used information based on open source intelligence, expert surveys and captive interviews). In addition to the ontology-based, abstract domain knowledge model, the proposed framework consists of an explorative information discovery functionality, which can couple concrete, case-related data from different knowledge bases with the abstract domain knowledge, to assist experts in the investigation of crimes and the discovery of new relations between different pieces of evidence. The proposed framework is illustrated using the exemplary use case scenario of Point-of-Sale (POS) Skimming. Furthermore, its flexibility, scalability and a potential integration into current and emerging police standards is discussed.