ESG information extraction with cross-sectoral and multi-source adaptation based on domain-tuned language models

Jaeyoung Lee, Misuk Kim

Abstract: Highlights•ESG is an indicator of a company’s nonfinancial performance.•Develop a model for extracting ESG-related information.•Building a dataset for ESG information extraction and classification.•Applying ESG-classifier to other sectors and multisource text data.•Proposal of additional construction of ESG dataset using pseudo labeling technique.