Project Details
Description
This project is funded by a Research Excellence Award (REA) for approximately £74,000 under the Strathclyde Research Studentship Scheme (SRSS).
Environmental, Social and Governance (ESG) credentials are of growing strategic importance for businesses, with ESG investing being increasingly prioritised by investors. Accurate ESG scoring of corporations is of paramount importance. ESG scoring systems in practice, however, have major weaknesses: (1) opaqueness in the proprietary methods used by private ESG scoring agencies; (2) observed inconsistencies in the ESG scores reported by alternative private ESG scoring systems (Berg et al., 2019); and (3) the extent of missing ESG scores across the population of corporations. Our proposed research directly addresses these weaknesses.
We seek to develop cutting-edge methods that offer transparency, consistency, and wide applicability in ESG measurement. We will leverage the power of Artificial Intelligence (AI), while addressing the ‘black-box' constraint of the underlying algorithms. State-of-the-art eXplainable Artificial Intelligence (XAI) techniques will be used, providing explainable outcomes. Such ‘white-box' XAI techniques will lead to transparent measurement of firms’ ESG performance, reconciliation of inconsistencies in existing ESG scoring systems, and wider application to the base of corporations.
The main contribution of our work will be the novel development and application of Natural Language Processing (NLP) based XAI approaches (Danilevsky et al., 2020) for textual-based measurement of ESG. Further contribution will be made by extending this work through the augmentation of textual analysis with audio characteristics of corporate earnings calls (such as manager pitch, tone and hesitancy), thus establishing an NLP classifier based on multiple modes of communication that may further enhance the reliability of our explainable ESG scoring approaches.
Environmental, Social and Governance (ESG) credentials are of growing strategic importance for businesses, with ESG investing being increasingly prioritised by investors. Accurate ESG scoring of corporations is of paramount importance. ESG scoring systems in practice, however, have major weaknesses: (1) opaqueness in the proprietary methods used by private ESG scoring agencies; (2) observed inconsistencies in the ESG scores reported by alternative private ESG scoring systems (Berg et al., 2019); and (3) the extent of missing ESG scores across the population of corporations. Our proposed research directly addresses these weaknesses.
We seek to develop cutting-edge methods that offer transparency, consistency, and wide applicability in ESG measurement. We will leverage the power of Artificial Intelligence (AI), while addressing the ‘black-box' constraint of the underlying algorithms. State-of-the-art eXplainable Artificial Intelligence (XAI) techniques will be used, providing explainable outcomes. Such ‘white-box' XAI techniques will lead to transparent measurement of firms’ ESG performance, reconciliation of inconsistencies in existing ESG scoring systems, and wider application to the base of corporations.
The main contribution of our work will be the novel development and application of Natural Language Processing (NLP) based XAI approaches (Danilevsky et al., 2020) for textual-based measurement of ESG. Further contribution will be made by extending this work through the augmentation of textual analysis with audio characteristics of corporate earnings calls (such as manager pitch, tone and hesitancy), thus establishing an NLP classifier based on multiple modes of communication that may further enhance the reliability of our explainable ESG scoring approaches.
Layman's description
We plan to use transparent and explainable AI techniques, applied to textual and audio content, to provide new insights into corporate Environmental, Social and Governance (ESG) performance.
Status | Active |
---|---|
Effective start/end date | 2/10/23 → 1/10/26 |
UN Sustainable Development Goals
In 2015, UN member states agreed to 17 global Sustainable Development Goals (SDGs) to end poverty, protect the planet and ensure prosperity for all. This project contributes towards the following SDG(s):
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.