My research is concerned with the problem of Ontology Matching. Ontology Matching (sometimes also refered to as Ontology Alignment) is about finding correspondences between elements (e.g. concepts or roles) of two ontologies covering the same or a similar domain. The matching process can be divided into two steps: First, a set of matching hypotheses is generated. Then, a subset of these hypotheses is chosen as final alignment. In my work I focus on several problems related to the second step. In particular, I suggest the use of logical reasoning to ensure the consistency of the final alignment.

If you are interested in this approach, I suggest to read a paper presented at the ESWC Phd-Symposium "The Relevance of Reasoning and Alignment Incoherence in Ontology Matching". It contains a nice introduction and overview over the work done so far.

My current work is best reflected in two papers: Look at the 2009 RR-paper, which describes a stable and efficient algorithm for incoherence based alignment debugging in detail. Optimal solutions to the debugging/extraction problems are proposed in the AAAI-2010 paper where Markov Logic is used (on top of Integer Linear Programming) resulting in a very flexible and powerful framework.

These days i'm a writing all the things down in my thesis, which will hopefully be finished in autumn 2010 ... It is now nearly one year later and I finally finished! You can download my thesis below. It is currently under review by Heiner Stuckenschmidt and Jerome Euzenat, which will hopefully accept it! You canl also download the software that I have been used in my thesis, called ALCOMO.