Abstract
This study aims to address the immature development of environmental pollution-related insurance in China by introducing machine learning technology to improve the traditional risk assessment model in order to scientifically and accurately quantify the pollution caused by enterprises in the production process. In this paper, public data such as annual reports and social responsibility reports of listed companies are collected, and the data on corporate emissions are identified by KMeans and GMM clustering, and then the probability of corporate illegal emissions behavior is predicted by using XGBoost classifier. At the same time, considering the environmental pollution risk of the location of the enterprise, Random Forest was used to predict the comprehensive risk and transfer probability. Finally, the enterprise discharge risk is evaluated by hierarchical analysis method. This study not only helps the enterprise's own risk management, but also provides a pricing basis for the insurance company and promotes the development of environmental pollution-related insurance in China, which has important theoretical and practical significance.
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