An Intelligent Risk Mining Method by Whole Process-Oriented Risk Analysis Model

This paper proposes a risk intelligent analysis method oriented to the whole process of events and applications, it is used to solve the problem of difficult and incomplete identification of risks, to dig out the risks behind it, and to provide help to ensure the safety of the public. Specifically, through step-by-step mining of incidents and application usage risks, combined with the event evolutionary graph, the model is run to calculate the similarity of text risks for comprehensive analysis, the innovative point is to propose a risk weighting model for the subject and object entities. Experiment shows that the method can reflect the complete risk of events and applications, and has improved recall and precision. The advantage of this method is the integration of multiple perspectives and disciplines, which has guiding significance in practice.