Last week saw the latest edition of the Semantics series of conferences, Semantics 2021, which took place in Amsterdam, Holland. Since 2005, the Semantics conferences provide the main European forum covering both advanced academic research in semantic technologies, as well as the state of the art in the relevant industrial applications.
Semantics 2021 saw a very strong presence of KMi researchers, both past and present!
Prof Enrico Motta gave the inaugural keynote of the conference, presenting a talk entitled “Combining Deep Learning and Common-Sense Reasoning in Visually Intelligent Robots”. The presentation centred on the PhD research carried out by Agnese Chiatti, a member of Enrico’s research group, who is investigating the use of common-sense reasoning to improve a robot’s ability to make sense of its environment – see https://robots.kmi.open.ac.uk for more details. This work is led by Agnese, with contributions coming also from Dr Enrico Daga and Dr Gianluca Bardaro.
Enrico’s keynote was followed by a session focusing on novel research advances related to the SPARQL query language and the first presentation in this session was given by Enrico Daga, who presented a paper entitled "Facade-X: an Opinionated Approach to SPARQL Anything". This contribution offered an original perspective to the problem of Knowledge Graph Construction. This work, which is supported by two EU grants, SPICE and Polifonia, was carried out in collaboration with another KMi-er, Dr Paul Mulholland, as well as researchers from the University of Bologna in Italy.
A few days later Enrico Daga gave another talk during the DBpedia Day at the conference, which was entitled "Capturing the semantics of documentary evidence of humanities research". The talk illustrated a hybrid method, combining machine learning and knowledge engineering, for identifying "themed" evidence in books, with case studies from the music domain. This line of research started during the Listening Experience Database (LED) project and is currently progressing in Polifonia.
Finally, the participation of KMi’s present and past research stars was completed by a keynote by Dr Vanessa Lopez, formerly a researcher in KMi and now at the IBM Research Centre in Dublin. Vanessa’s presentation, entitled “Knowledge Graphs for Social Good: Protecting Vital Health and Social Programs”, focused on the problem of realising computer programs that can correctly implement government policy, e.g., in the provision of healthcare services.