Development of a semantic representation model of criminal information to support the assessment of risk situations

• Grant number: 18/01035-6
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: July 01, 2018
• End date: December 31, 2018
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Leonardo Castro Botega
• Grantee: Jordan Ferreira Saran
• Host Institution: Centro Universitário Eurípedes Soares da Rocha. Fundação de Ensino Eurípedes Soares Rocha (FEESR). Marília , SP, Brazil

Abstract

Situational Awareness (SAW) refers to the level of consciousness that an individual or team holds over a situation. In the area of risk management and criminal data analysis, SAW failures can induce human operators to make mistakes in decision making and pose risks to life or property. In this context, risk assessment processes, which commonly involve data mining, fusion and other methods, present opportunities to generate better information and contribute to the improvement of the SAW of crime and risk analysts. However, the characterization of complex scenarios is subject to problems of representation and expressiveness of the information, which may influence its interpretation due to their quality and significance, generating uncertainties. The state-of-the-art in representation of information on risk situations and related areas presents approaches with limited use of information quality. In addition, the solutions are restricted to syntactic mechanisms for the determination of relations between information, negatively restricting the assertiveness of the results. Thus, this proposal aims to develop a new approach to semantic representation of information of risk situations, more specifically creating domain ontologies, instantiated with crime data and information quality. In a case study, real information on crimes, represented by the new semantic model and consumed by computational inference processes, will be processed, aiming to characterize robbery and theft situations. To evaluate the approach, the information represented will be analyzed by experts and scored on the level of SAW provided. (AU)