Credit Evaluation Model of Small and Micro Enterprises' Financing Risk Based on Modified BP Neural Network

摘要

With the development of the Internet, small and micro enterprises have developed rapidly and have become an important driving force for social and economic development. Therefore, it is particularly important to establish a sound financing risk evaluation system. This article is based on the big data analysis environment, combined with the direction of financial big data and data mining. The BP neural network model is transformed, and the loss function is transformed from the basic cross-entropy form to the Focal Loss function that can control the weight of positive and negative samples. Such improvements make the evaluation model more suitable for the characteristics of indicators and the actual development of small and micro enterprises. The small and micro enterprise financing risk reputation evaluation model established in this paper can be used as reference information for financial institutions to invest in small and micro enterprises, and promote the establishment of a good credit cooperation relationship between financial institutions and small and micro enterprises.