NLP for SDGs: Measuring Corporate Alignment with the Sustainable Development Goals

This study uses advanced natural language processing methods to identify companies that are aligned with the UN SDGs based on the text in their sustainability disclosures. Using CSR reports of Russel 1000 companies between 2010-2019 we apply a logistic classifier, support vector machines (SVM) and a fully-connected neural network to predict alignment with the SDGs. Specifically, we use word embeddings to augment dictionary-based input features as well as use the embeddings as features themselves based on Word2Vec and Doc2Vec models to classify companies’ alignment with the SDGs over time. Particularly Doc2Vec embedding inputs to the SVM classifier result in high average accuracy of above 80% for predicting alignment.