projects by Matteus Tanha
Question Answering with Deep Learning
In this liveProject, you’ll step into the role of a data scientist working for an investment firm. Your company wants to make sure their investments meet European Union guidelines for environmental sustainability. That’s where you come in.
The EU taxonomy for sustainable finance is big, complex, and confusing. Your bosses need a program that saves them from searching through hundreds of pages whenever they have a query. You’ve been tasked with building a machine learning model that can pose certain questions to the EU guidelines, and return reliable answers.
Your challenges will include extracting text data from the EU taxonomy document, and matching environment questions with the corresponding paragraph in the guidelines. You’ll then set up a pretrained transformer Question-Answering model, evaluate its performance, and combine it with your question-paragraph model for an end-to-end solution. When you’re done, you’ll have an interface into which you can type a sustainable finance question and receive the correct answer from the EU guidelines.
