ESG2PreEM: Automated ESG grade assessment framework using pre-trained ensemble models.

Advanced Search

Haein Lee, Seon Hong Lee, Heungju Park, Jang Hyun Kim, Hae Sun Jung

Author Information

Haein Lee: Department of Applied Artificial Intelligence/ Department of Human Artificial Intelligence Interaction, Sungkyunkwan University, 03063, Seoul, South Korea.
Seon Hong Lee: Department of Applied Artificial Intelligence/ Department of Human Artificial Intelligence Interaction, Sungkyunkwan University, 03063, Seoul, South Korea.
Heungju Park: SKK Business School, Sungkyunkwan University, 03063, Seoul, South Korea.
Jang Hyun Kim: Department of Interaction Science/ Department of Human Artificial Intelligence Interaction, Sungkyunkwan University, 03063, Seoul, South Korea.
Hae Sun Jung: Department of Applied Artificial Intelligence, Sungkyunkwan University, 03063, Seoul, South Korea.

PMID: 38404885 DOI: 10.1016/j.heliyon.2024.e26404

Incorporating environmental, social, and governance (ESG) criteria is essential for promoting sustainability in business and is considered a set of principles that can increase a firm's value. This research proposes a strategy using text-based automated techniques to rate ESG. For autonomous classification, data were collected from the news archive LexisNexis and classified as E, S, or G based on the ESG materials provided by the Refinitiv-Sustainable Leadership Monitor, which has over 450 metrics. In addition, Bidirectional Encoder Representations from Transformers (BERT), Robustly optimized BERT approach (RoBERTa), and A Lite BERT (ALBERT) models were trained to accurately categorize preprocessed ESG documents using a voting ensemble model, and their performances were measured. The accuracy of the ensemble model utilizing BERT and ALBERT was found to be 80.79% with batch size 20. Additionally, this research validated the performance of the framework for companies included in the Dow Jones Industrial Average (DJIA) and compared it with the grade provided by Morgan Stanley Capital International (MSCI), a globally renowned ESG rating agency known for having the highest creditworthiness. This study supports the use of sophisticated natural language processing (NLP) techniques to attain important knowledge from large amounts of text-based data to improve ESG assessment criteria established by different rating agencies.

BERT ESG Ensemble Natural language processing (NLP) Pretrained language model

Financ Res Lett. 2021 Jan;38:101870 [PMID: 36569646]
Educ Psychol Meas. 2023 Aug;83(4):831-854 [PMID: 37398846]
Math Biosci Eng. 2023 Aug 29;20(9):17018-17036 [PMID: 37920045]

Journal Article

Efficient brain tumor grade classification using ensemble deep learning models.Ensemble fusion model for improved lung abnormality classification: Leveraging pre-trained models.Pre-trained models, data augmentation, and ensemble learning for biomedical information extraction and document classification.Automatic Detection of Liver Cancer Using Hybrid Pre-Trained Models.Relation Extraction from Clinical Narratives Using Pre-trained Language Models.Using hybrid pre-trained models for breast cancer detection.Automated Cardiovascular Pathology Assessment Using Semantic Segmentation and Ensemble Learning.Lottery Jackpots Exist in Pre-Trained Models.Multi-domain Urdu fake news detection using pre-trained ensemble model.Automated detection of diabetic subject using pre-trained 2D-CNN models with frequency spectrum images extracted from heart rate signals.

Refining the prediction of user satisfaction on chat-based AI applications with unsupervised filtering of rating text inconsistencies.Deep-learning-based stock market prediction incorporating ESG sentiment and technical indicators.Decoding Bitcoin: leveraging macro- and micro-factors in time series analysis for price prediction.Expansive data, extensive model: Investigating discussion topics around LLM through unsupervised machine learning in academic papers and news.

OpenLB
Open Library of Bioscience