UPM|Mediateca

Enrico Motta, Open University. Symposium Artificial Intelligence and Ceremony of Awards. 21 and 22 October 2019. Over the past 15-20 years we have witnessed a paradigm shift in Computer Science, brought about by the unprecedented availability of large-scale amounts of data. As a result of this paradigm shift new exciting opportunities have arisen, allowing us to study the dynamics of a variety of phenomena, such as social networks, people’s mobility patterns, shopping behaviours, etc. Within this overall context, a particularly interesting area of analysis is that of scholarly data, which allows us to gain significant insights about the dynamics of the research world. These insights are important not just because they allow us to better understand, at different levels of granularity, how research evolves, but also because they can help us to identify biases and distortions in the system of scientific research and inform corrective actions. In my presentation I will illustrate our work in this area, which has led to a number of significant research outcomes as well as the deployment of innovative technologies in commercial settings. Highlights include: an innovative hybrid approach to the generation of highly detailed maps of the space of research topics in a given field of science. a novel solution, inspired by epistemological models of science, which is able to predict the emergence of a new research field before this has been explicitly recognised by a research community. a geo-political analysis of Computer Science conferences that indicates a rather static landscape in which, despite the massive increase in scholarly production, new entries struggle to emerge. and a novel solution, in routine use at Springer Nature since 2016, which automatically generates the scholarly metadata for all conference proceedings in Computer Science. This has led to substantial efficiency improvements in the editorial workflow, as well as several million additional downloads of conference proceedings on the Springer portal.