Abstract: Set pair analysis theory(SPA) is a quantitative analysis method for dealing with uncertainty and uncertainty. In the paper, FSPA algorithm is established by using the fuzzy connection degree of the SPA to further explore and analyze the difference coefficient. The aim is to discuss the close degree between the class and the adjacent categories by using the fuzzy relation data mining, and to discuss similarity and difference information. In order to verify the applicability of the model, the risk assessment was carried out by measuring the basic data of six debris flows in Fangshan District and Mentougou area of Beijing, and the evaluation results were compared with the theory of extension theory. the study concluded that Fuzzy set pair analysis theory can depth mining similarities and differences between sample types and adjacent categories inverse relation. By comparing with the field survey data, It was found that FSPA evaluation results of information mining more realistic. The evaluation results of this method are consistent with the results of the extension evaluation. The method has high accuracy and precision.