Research On Evaluation Of Small And Medium-sized Enterprises In Information Technology Industry

As an indispensable and important economy in economic development,SMEs(small and medium-sized enterprises)are facing huge challenges in their survival and development.The most critical challenges are the difficulty and high cost of financing.In recent years,the credit risk assessment of SMEs has been greatly developed in terms of assessment indicators,assessment methods,assessment scope and financial model.However,banks and other financial institutions still face huge challenges and threats when assessing credit risk,which directly affects the financing capacity of SMEs.Therefore,improving the credit evaluation level of SMEs will effectively improve the financing difficulties of SMEs.Starting from the industry,this paper carries out the research on credit evaluation of SMEs facing the information technology industry,mainly including three aspects.First,it discusses the characteristics of SMEs in the information technology industry and the causes of credit risk.First of all,it describes the development status of SMEs in China,including the problems of SMEs in terms of enterprise scale,business management,financing environment,etc;Secondly,it introduces the development characteristics of the information technology industry,including the definition,characteristics,development opportunities and challenges of the information technology industry;Finally,it introduces the relevant content of credit risk of SMEs,including the definition,formation mechanism and management of credit and credit risk.The second is to build a credit evaluation index system for SMEs in the information technology industry.Firstly,according to the characteristics of SMEs in the information technology industry and the relevant principles of indicator selection,the sample data is selected;Secondly,the sample data is preprocessed,including: preliminary screening of indicators,filling of missing values in the sample data,data standardization,sample balance,etc;Finally,the final indicator system is determined according to different evaluation models.The third is to build a credit evaluation model for SMEs in the information technology industry.Firstly,the credit evaluation model is constructed based on the logistic regression(LR)and support vector machine(SVM);Then the LR model and SVM model are evaluated by using the confusion matrix;Then the significance between variables and labels is tested by F-test,and the variables with statistical significance are analyzed;Finally,the model evaluation results of RF,MLP,LR and SVM are compared.The empirical results show that: first,the selection of indicators is more targeted and the accuracy of the credit evaluation model is higher when carrying out the research on the credit evaluation of SMEs in the information technology industry with the industry as the research object;Second,the accuracy of the credit evaluation model based on SVM and LR is better than that of RF and MLP,and SVM is the best model among them;The third is to analyze the variables that affect credit risk,among which the proportion of variable R&D expenses has a significant correlation with whether or not to default.The research in this paper is helpful to enrich the research on credit evaluation of SMEs based on industry,and has certain reference value for carrying out credit evaluation of SMEs in the actual information technology industry.