Research Article
Analysis of the Influential Factors and Prediction of Corporate ESG Performance under Multi-source Data Fusion - Based on Frontier Machine Learning
@INPROCEEDINGS{10.4108/eai.2-12-2022.2328731, author={Chenhong Zheng and Mengqian Zhang and Cong Zeng and Fangshun Xiao and Mengzhe Liu}, title={Analysis of the Influential Factors and Prediction of Corporate ESG Performance under Multi-source Data Fusion - Based on Frontier Machine Learning}, proceedings={Proceedings of the 3rd International Conference on Big Data Economy and Information Management, BDEIM 2022, December 2-3, 2022, Zhengzhou, China}, publisher={EAI}, proceedings_a={BDEIM}, year={2023}, month={6}, keywords={esg; machine learning; bp; random forest; svm}, doi={10.4108/eai.2-12-2022.2328731} }- Chenhong Zheng
Mengqian Zhang
Cong Zeng
Fangshun Xiao
Mengzhe Liu
Year: 2023
Analysis of the Influential Factors and Prediction of Corporate ESG Performance under Multi-source Data Fusion - Based on Frontier Machine Learning
BDEIM
EAI
DOI: 10.4108/eai.2-12-2022.2328731
Abstract
Environmental, Social and Governance (ESG) performance, which reflect the degree of corporate green development and sustainable development potential, is one of the key factors determining the long-term development of enterprises. Based on the integration of macro data and enterprise data, this paper firstly explores the factors influencing ESG performance from the perspective of both external environment and internal characteris-tics of enterprises using the Lasso method, and then screens them effectively. Then, based on the screening results of the influencing factors and the underlying data set, var-ious machine learning methods such as BP, Random Forest, KNN and SVM models are used to predict the ESG performance of enterprises, and the prediction accuracy of each method is compared and analyzed. Finally, based on the prediction results and the screening results, the importance of each influencing factor on the ESG performance of enterprises is ranked comprehensively.