The ZBW provided us with a dataset of around 7.000 text files of studies from the EconStore Database. We were using machine learning based question and answering systems with the Haystack Framework to extract information from scientific studies. e.g. "what datasets were used in this study?" or "who funded this study?" We also built a React Frontend to display the newly extracted meta-data and make our new database seachable.
Log in or sign up for Devpost to join the conversation.