ESG taxonomy is a comprehensive, standardized framework for assessing and categorizing environmental, social, and governance (ESG) factors in the investment process. It is designed to provide a common language for investors, companies, and other stakeholders to understand and assess the impact of ESG factors on the financial performance of companies and investments.
Permutable strives to provide the best in ESG taxonomy to allow for full extraction of datapoints from global public data-sources. We use over 1000+ markers to label ESG data. Below are a sample of markers we use on our datasets.
Our methodology involves a rigorous assessment process that scans 500,000 articles globally on a daily basis. Using advanced AI algorithms and data analytics we deploy a multi-step process for evaluating sentiment towards ESG factors. Initially, unique entities such as organizations are identified using machine learning models. Ontologies comprising keywords and phrases for each ESG topic and subtopic are defined and then applied to the text.
Subsequently, sentiment generation occurs by submitting the text, recognized entities, and known subtopics to a sentiment model, producing classifications as positive or negative. Monitoring sentiment signals over time, recurrent and infrequent patterns, such as negative story breakouts, are identified using historical data.