Abstract
Through in-depth analysis of credit risk in the context of the development of Internet finance, the study recognized that its diversification and complexity not only include financial risks, but also involve the impact of non-financial factors. Based on this, combined with ESG factors, principal component analysis and grey correlation method are used to optimize the indicators, ensuring the comprehensiveness and accuracy of the indicator system. Using Convolutional Neural Network (CNN) as the warning model, considering the dynamic and static characteristics of data, a sub convolutional network was designed for training different types of data. At the same time, the credit rating division and warning threshold selection methods were optimized, improving the accuracy and practicality of the model. Through experimental verification, we found that the CNN model has a high accuracy in credit risk warning, and it also shows excellent classification performance and comprehensive effect compared to other models. Through systematic research and application, this paper provides a complete solution for the credit risk management of Internet financial enterprises, and makes positive contributions to the stability of the financial system and risk prevention. At the same time, we also emphasize the important role of technological innovation in financial intelligent management, especially the application of innovation systems centered on AI and knowledge graphs in the field of credit management, which points out the direction for the future development of the financial industry.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call