It will do so by developing a robust conceptual model addressing all key stages of the innovation process; mining large volumes of data on research results and impacts; and analysis of these data using topic modelling, machine learning and other techniques aimed at natural language processing. The Data4Impact consortium possesses specialist knowledge of the health domain & indicator systems, and is uniquely placed to mine data and apply big data approaches thanks to the partner’s long-standing involvement in OA e-infrastructures and big data analytics.