Research On Credit Evaluation Of Small And Micro Enterprises Based On Paraconsistent Soft Sets Under Incomplete And Inconsistent Data

In recent years,the overall scale of small and micro enterprises is growing day by day.And the production capacity and economic benefits brought by them,which plays an important role in supporting China’s national economy,can not be ignored.However,due to the small scale of a single small and micro enterprise,financial institutions lack the ability to control the default risk of small and micro enterprises,which leads to that expensive and difficult financing becoming the main factors restricting the development of small and micro enterprises.And the key to solve this problem is to build a suitable credit evaluation algorithm for small and micro enterprises.Under the big data environment,the credit data of small and micro enterprises have different characteristics from traditional credit data,such as poor(incomplete and inconsistent),dynamic and multi-dimensional.On the one hand,incomplete data are produced because that the data collection institutions cover different users and the data collection is not standardized enough.On the other hand,due to the different sources and collection time of credit data,or the loopholes in a certain link when collecting,calling and storing multi-source data,inconsistent data are generated.However,in most cases,incomplete and inconsistent data exist simultaneously.Based on this,this study takes the sub-coordinated soft set with the advantage of big data analysis as the theoretical basis.It is should be noted that,in the big data environment,the credit data of small and micro enterprises not only has poor quality(incomplete and inconsistent),but also changes with the passage of time,showing dynamic.Meanwhile,the data often involve various dimensions and show multi-dimensionality.In order to solve the credit evaluation problem of small and micro enterprises involving dynamic incomplete and inconsistent data,this study constructs a credit evaluation algorithm of small and micro enterprises based on dynamic sub-coordinated soft sets.Additionally,in order to solve the problem of credit evaluation of small and micro enterprises involving multi-dimensional incomplete and inconsistent data,this study also constructs a credit evaluation algorithm of small and micro enterprises based on Q-sub-coordinated soft sets,aiming to provide scientific guidance for credit evaluation of small and micro enterprises under big data environment.The specific research contents of this paper are as follows:In view of the poor(incomplete and inconsistent)and dynamic characteristics of the credit data of small and micro enterprises,this paper introduces the time factor to further extends the sub-coordinated soft set to the dynamic sub-coordinated soft set,and defines some basic operations of the dynamic sub-coordinated soft set.On this basis,this paper puts forward a dynamic sub-coordinated soft decision-making system,and constructs a credit evaluation algorithm for small and micro enterprises based on dynamic sub-coordination soft set,which can deal with the problems including dynamic incomplete and inconsistent data.In addition,the algorithm is applied to a practical loan problem of small and micro enterprises,and sensitivity analysis and comparative analysis are carried out,which proves the effectiveness and feasibility of the algorithm.In view of the poor(incomplete and inconsistent)and multi-dimensional characteristics of the credit data of small and micro enterprises,this paper further expands the sub-coordinated soft set to Q-sub-coordinated soft set,which not only retains the advantages of the former,but also can handle two-dimensional data.This paper introduces the priority parameter set Q,puts forward the definition of Qsub-coordinated soft set,and discusses its related operations.Finally,a credit evaluation algorithm based on Q-sub-coordinated soft set is constructed,and the algorithm is applied to the problem of what kind of small and micro enterprises to invest in,develop and build.Meanwhile,comparative analysis is presented to prove the superiority and effectiveness of the proposed algorithm.