Leveraging artificial intelligence to improve Environmental, Social, and Governance (ESG) data and assess the quality of ESG reports

利用人工智能改进环境、社会和治理 (ESG) 数据并评估 ESG 报告的质量

• 批准号: 571240-2022
• 负责人: Morales, Manuel
• 金额: $ 1.46万
• 依托单位: Université de Montréal
• 依托单位国家: 加拿大
• 项目类别: Idea to Innovation
• 财政年份: 2021
• 资助国家: 加拿大
• 起止时间: 2021-01-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=729094
• 关键词: Leveraging; artificial; intelligence; improve; Environmental

Most of the investors incorporate ESG scores in their portfolio and rely on ESG rating providers. However, there is a lack of consistency in the ESG impact measurements as the rating companies develop their own methodologies and indicators which are often different from other providers. A study shows that there is only 0.54 of correlation between six major ESG rating providers. This brings challenges for investors to make an investment decision based on the ESG ratings as their decisions may vary based on the rating agency. The lack of methodological standard for measuring ESG performance, poses problems not only for investors but also for rating agencies to produce reliable, consistent, and transparent ESG scores.The information necessary to understand the ESG impact of a company is in an unstructured relational format and artificial intelligence (AI) can be used in these efforts. In the traditional manner, extracting insights from these unstructured and qualitative text data, which are social media, companies' reports, earning calls, and news, has been done manually. In our product, we employ AI to effectively automate this extraction and provide more efficient and transparent large-scale analysis. More specifically, we use natural language processing (NLP) to capture the sentiment, entities, information, and semantics from text data and transform the data to interpretable structured data and information. Moreover, we leverage AI to assess the large-scale analysis of climate-related financial disclosures automatically.

大多数投资者都将ESG分数纳入其投资组合中，并依靠ESG评级提供商。但是，由于评级公司开发自己的方法和指标通常与其他提供商不同，因此ESG影响测量缺乏一致性。一项研究表明，六个主要的ESG评级提供商之间只有0.54个相关性。这给投资者带来了基于ESG评级做出投资决定的挑战，因为他们的决策可能会根据评级机构而有所不同。缺乏用于衡量ESG性能的方法论标准，不仅对投资者构成了问题，而且对评级机构产生可靠，一致和透明的ESG分数。了解公司的ESG影响所必需的信息是在这些工作中以非结构化的关系格式和人工智能（AI）的使用。以传统的方式，手动完成了这些非结构化和定性文本数据的见解，这些文本数据是社交媒体，公司的报告，赚取的电话和新闻。在我们的产品中，我们采用AI有效地自动化此提取，并提供更有效和透明的大规模分析。更具体地说，我们使用自然语言处理（NLP）从文本数据中捕获情感，实体，信息和语义，并将数据转换为可解释的结构化数据和信息。此外，我们利用AI自动评估对气候相关财务披露的大规模分析。