Article Preview
Top1. Introduction
With the rapid development of economy, problems such as population expansion, excessive use of resources, and environmental pollution continue to emerge (Song 2019). With the increasing demand of human consumption, in order to meet the expanding consumption demand, there are predatory exploitation and destruction of natural resources, which have caused serious environmental pollution and ecological crisis (Zou 2019). Only when the inappropriate consumption of human beings is fully curbed, and people take the road of sustainable consumption, can the environment be protected and the ecological crisis can be gradually eliminated (Bengtsson 2018). At present, sustainable development has become a globally recognized economic development strategy (Barrow 2018). The two basic aspects of sustainable development are sustainable production and sustainable consumption (Lukman 2016). Among them, sustainable consumption means that contemporary people cannot exceed the limit of ecological environment carrying capacity when meeting the needs of their consumption development. Consumption should be conducive to environmental protection and ecological balance (Govindan 2018). It requires not only the optimal and sustainable utilization of resources, but also the minimum discharge of waste and the minimum pollution to the environment (Wang 2019). In order to realize the concept of sustainable development, the sharing economy model appears in the existing market. Sharing economy is a new economic model which optimizes resource allocation and efficient social governance. It is based on the Internet and other modern information technology support. The resource supplier will provide the temporarily idle resources to the resource demander with compensation through the technology platform. The demand side obtains the right to use resources, while the supply side gets the corresponding reward (Puschmann 2016). However, under the background of sharing economy, the alienation of platform social responsibility behavior has cast a shadow on the promotion of sustainable consumption, which is mainly reflected in the following aspects: the content of social responsibility of platform enterprises needs to be defined; the alienation of social responsibility needs to be addressed; the behavior of consumer dishonesty in the level of credit environment needs to be restricted; the overall credit environment needs to be improved; under the level of consumer rights and interests, the rights and interests of consumers in the sharing platform need to be protected; the governance system of social responsibility in platform enterprises needs to be established urgently (Vith 2019). Among them, the most important is to establish a new enterprise credit risk assessment model with high accuracy under the sharing economy (Richter 2017).
At present, there are few researches on enterprise credit risk assessment model, and most of them focus on the establishment of risk assessment system. Among them, established a risk index system by using the analytic hierarchy process (AHP) according to the characteristics of credit business process of small and medium-sized enterprises of commercial banks, and carried out risk analysis. On the basis of risk identification, the fuzzy evaluation model and judgment matrix were established to effectively reduce the enterprise credit risk (Shi 2016); in order to explore enterprise credit risk assessment, compared the application effect of several common neural network models in China's SME data sets, and found that probabilistic neural network has the minimum error rate, the highest area under curve (AUC) value, and has good robustness (Huang 2018); used the nonlinear least squares support vector machine (LS-SVM) model to analyze the credit risk indexes of supply chain enterprises, and combined with the index selection principle to determine the final index system, and constructed the enterprise online supply chain financial credit index. The results show that the classification accuracy of LS-SVM evaluation model is higher than that of Logistic regression model, and has strong generalization ability (Wang 2019); proposed an enterprise credit risk assessment method based on improved genetic algorithm, established a grid structure of longitude and latitude, improved and optimized longitude latitude grid genetic algorithm, and improved enterprise credit risk assessment. This method is better than traditional strategies and can be widely used (Yang 2020). It can be seen that for the analysis of enterprise credit rating, AHP method and traditional data analysis model are often used. This method has poor accuracy and is difficult to evaluate enterprise credit effectively.