Leveraging BERT Language Models for Multi-Lingual ESG Issue Identification

Trading Central Labs partook in the Multi-Lingual ESG Issue Identification project, aiming to classify financial documents into ESG issue labels. The team proposed the use of multiple BERT-based models, for ESG issue classification. Their RoBERTa classifier ranked second in the English category and showed promising results in French. Their SVM-based model also secured second place in Chinese, showcasing the potential of their models for accurate cross-lingual ESG issue classification.