Beyond syntactic data fusion in the context of criminal data analysis

Situational Awareness (SAW) refers to the level of consciousness that an individual or team holds about a situation. In the area of risk management, SAW failures can induce human operators to make mistakes in the decision-making process and to risk life, property or the environment. Data fusion processes present opportunities to improve the SAW of operators of risk assessment systems. However, dynamic scenarios are subject to uncertainties, especially when the data are derived from human intelligence (HUMINT). The state of the art in data and information fusion techniques presents approaches restricted to syntactic and/or pragmatic mechanisms for determining synergy between information, negatively influencing the results of the fusion. This paper proposes to demonstrate in practice such limitations and present an approach toward the use of semantic models of representation and information inference, using the domain of risk management as a case study, more specifically with criminal data from Brazil.