Automatic Extraction of Sustainability Indicators
BBVA, a banking giant with strong influence in Spain, Turkey, and Latin America, faces the daily challenge of evaluating the sustainability of companies through ESG (Environmental, Social, and Governance) indicators for granting loans. This process is complex and time-consuming due to the need to analyze extensive annual reports full of disorganized information. To overcome this obstacle and optimize its processes with advanced technology, BBVA needed a solution that would effectively automate the extraction of these indicators.
At WhiteBox, we quickly identified that Large-Scale Language Models (LLMs) were the ideal solution to address this challenge. We developed a pioneering application that uses the LangChain framework and OpenAI LLM technology (GPT-4), designed to extract ESG indicators from a set of documents with precision and efficiency. We complement this innovation with an interactive visualization tool, Microsoft Power BI, which allows a dynamic exploration of the extracted data.
The implementation of our solution for the automatic extraction of ESG indicators has changed the way in which BBVA approaches the analysis of the corporate sustainability of its clients. Now, the bank can prioritize financing sustainable businesses without investing the valuable time of financial analysts in manually reviewing documents. Quick access to ESG indicators through an interactive dashboard has not only optimized internal processes but also generated significant savings in time and resources.
Data Engineering Implant
Data Science Implant
