Extended automatic reasoning of criminal investigation

This paper proposes an extended fuzzy reasoning principle (EFRP) for criminal investigation and a new framework for automatic reasoning based on extension relationship mapping. We develop a conceptual model of the cognitive abilities that a criminal investigator uses during a criminal investigation. Two mapping models are set up by constructing a pattern on practical crime model and on-the-spot model respectively. Research has shown that the premise of automatic reasoning is to set up patterns of case scene and case-cracking clue; moreover, the complementary of the clues is also key point. The viewpoint of authors is that the reliability of fuzzy automatic reasoning depends on man-computer interaction results. Simultaneously, choosing a case-cracking clue should be determined by comprehensive evaluations, self-learning, intuitional and logical judgments.