Identifying Environmental Information Disclosure Manipulation Behavior through Machine Learning: A Comparative Analysis of Recognition Models
Zhongqi Xu,Yuanzhe Li,Yuxiang Hu,Hengli Chen,Chengmeng Wu,Yan Wang
DOI: https://doi.org/10.1109/ICISCAE59047.2023.10393564
2023-09-23
Abstract:Manipulation of corporate environmental information disclosure is highly covert, posing significant challenges to the identification and judgment of such manipulative behaviors. In this study, we apply machine learning methods to construct a model for identifying manipulative behaviors in environmental information disclosure. We collect environmental pollution penalty cases of Chinese listed companies from 2011 to 2020 and combine them with the "pressure pool" index system to build training and testing datasets. We compare the recognition abilities of logistic regression, decision tree, and random forest models. To address the issue of data imbalance, we compare two oversampling techniques, Borderline-SMOTE and ADASYN. In the final detection results, the Borderline-SMOTE algorithm demonstrates better identification performance, and the Borderline-SMOTE-RF model outperforms the logistic regression and decision tree models. We hope that our research can provide references for regulatory authorities, accelerate the improvement of mandatory environmental information disclosure regulations for listed companies, enhance the detection, identification, and early warning capabilities of manipulative behaviors in environmental information disclosure, and promote the quality of environmental information disclosure.
Computer Science,Business,Environmental Science