Big Data approaches to estimating the impact of EU research funding on innovation development

Abstract

EU will spend around € 80 billion in supporting research through H2020 instrument alone. Such significant amounts of money dedicated to supporting research projects naturally beg the question: What impacts resulted from this EU funding? Answering this question is hard, especially keeping in mind that EU-funded researchers already struggle with significant amounts of paperwork and are burdened by various periodic surveys. In this paper, we propose a method to estimate the impact of EU funding on innovation development using non-invasive data acquisition methods, which do not further contribute to the EU survey fatigue. The method utilizes supervised machine-learning techniques to infer data on the innovation outputs of the EU-funded companies from publicly available data on their websites. Gathered information is then compared to the control group of enterprises from the same geographical area and economic sector. The panel of target and control group of enterprises is analyzed through time to uncover potential lagged/ long-term effects of the EU funding. The paper is structured as follows: the first part presents an overview of the need to estimate the impact of EU (research) funding and the methodological challenges associated with it. Second part presents the methodology and study design. The third part presents the preliminary results and their implications, while the forth – discusses outstanding challenges and the next steps.