Boris Motik from University of Oxford, United Kingdom, has been awarded the 2007 Cor Baayen Award for a most promising young researcher in computer science and applied mathematics by ERCIM.
In a tight competition with 17 finalists, ERCIM has awarded Boris Motik for the outstanding quality of his work concerning reasoning algorithms and systems for Description Logics (DLs) - a family of knowledge representation formalisms with applications in numerous areas of computer science. DLs provide the basis for the Web Ontology Language (OWL) - the ontology language defined by the World Wide Web Consortium (W3C) that has become the de facto standard for ontology development in fields as diverse as geography, geology, astronomy, agriculture and the life sciences. Boris has already made wide ranging contributions to research, including both new theoretical results, and practical systems that promise to change our notion of tractability in ontology reasoning.
Boris's output is impressive: in spite of the fact that he only completed his PhD in 2006, he has already published more than 40 articles in leading international conferences and journals, and won the best paper prize at the 2005 International Semantic Web Conference (ISWC 2005) for his paper "On the Properties of Metamodeling in OWL".
Boris studied for his Bachelor's and Master's degrees at the Faculty of Electrical Engineering and Computing at University of Zagreb, Croatia, where he received the faculty award Josip LonÄ?ar for the best student of the generation. Boris's PhD work at Technical University Karlsruhe, Germany, broke new ground in scalable reasoning for ontology languages. Boris showed that the data complexity (i.e., the complexity measured in the size of data only) of ontology reasoning is lower than the combined complexity (i.e., the complexity measured in the size of data and the ontology) of reasoning. This shows that there are no fundamental limitations to scalability of reasoning in data intensive application -- that is, in applications where the size of the data dominates the size of the ontology. He also developed a completely new DL reasoning technique based on a reduction to disjunctive datalog. Unlike the existing techniques, this new technique exhibits worst-case optimal complexity, and it also enables the reuse of established optimisation techniques, such as magic sets, that greatly improve performance of reasoning with large amounts of data.
Boris's contributions were not only theoretical: he implemented a new reasoning system (KAON2) based on his reasoning algorithms and demonstrated that, on relatively simple ontologies with large data sets, the new system can answer queries several orders of magnitude faster than existing tableau based implementations. This work is of great importance to the future of the Semantic Web and ontology language applications more generally, where an inability to deal with large volumes of data has been a serious impediment to wider adoption. This is illustrated by the fact that Boris's work is already the basis for a commercial ontology management system being marketed by ontoprise GmbH.
In addition to his work on KAON2, Boris has made several other important research contributions. In his work on meta-modelling, he showed that the basic reasoning problems become undecidable if certain features of the Resource Description Framework (RDF) are combined with more expressive languages such as OWL, and that decidability can be restored by restricting the use of syntax reflection. This is an important result as it shows that it is possible for ontology languages to support meta-modelling while still retaining decidability, thus addressing the requirements of many Semantic Web applications. Boris's work on combining the best features of the Description Logic and Logic Programming paradigms has also been highly influential. He also worked on bridging the gap between description logics and relational databases, and proposed a framework for the integration of the open-world semantics of the former and the closed-world semantics of the latter formalism. Finally, he developed a novel calculus for DL reasoning based on hypertableau, and implemented it in a new reasoning system HermiT. This system was the first one to process certain large life-science terminologies, thus solving some long-standing open problems in scalability of reasoning with large terminologies.
17 finalists for the 2007 Cor Baayen Award have been nominated by the ERCIM institutes:
Marcin Bienkowski, Poland
Yves Bontemps, Belgium
Nick Cook, United Kingdom
Henri Dubois-Ferriere, Switzerland
Felix Klaedtke, Germany
Nikos Komodakis, Greece
Levente Kovacs, Hungary
DÃ¡niel Marx, Hungary
Boris Motik, United Kingdom
Guillaume Perrin, France
Srikumar Ramalingam, France
Andrey Rybalchenko ,Germany
Panu KorpipÃ¤Ã¤, Finland
Magnus Sahlgren, Sweden
Risto Sarvas, Finland
Elias Tsigaridas, Greece
Danny Weyns, Belgium
According to the award rules, each institute was allowed to select up to two finalists from its country.