Assessment of the European mobility research landscape to support policy shaping through artificial intelligence models

This paper presents an approach for assessing EU-funded mobility research initiatives that relies on natural language processing (NLP) techniques. The developed prototype acts as a digital assistant that helps to analyze the mobility research landscape and delivers a bird-eye view of its status, gaps, and bottlenecks. We present data-based models that exploit common NLP techniques used for topic modeling and information retrieval to automatize the analysis of the textual data of over 40,000 H2020 and PF7 research projects and to deliver a series of metrics that support insight discovery. Further, we present an open-access dashboard that visually inspects the model results. Based on the developed models, we provide high-level strategic recommendations for future mobility development. A particular use case focuses on digitalization in mobility.