With the emergence of semantic web and linked data technologies, as well as initiatives around government open data, the amount of data available and interconnected on the web has increased dramatically in the last few years. However, even if taking each source individually, our ability to make use of these data and extract meaningful and useful information out of them is already questionable. Putting data on the web now also means that we are confronted with increasing amounts of noise, lack of context, lack of meaning, etc. In this talk, I will present a set of techniques and approaches we have been experimenting with which try to make sense of such web data, borrowing from the fields of knowledge discovery, data mining and reasoning to embrace their heterogeneity. I will show examples of this approach in two areas that, increasingly, require data to be sourced and interconnected from a large variety of distributed providers: The booming field of (open) education, and the ever more critical one of personal information management.
Mathieu d'Aquin is a research fellow at the Knowledge Media institute (KMi) of the Open University in Milton Keynes, UK. His main focus is on methods and tools to build intelligent applications relying on formalised knowledge distributed online. As part of several projects, Mathieu has worked on many aspects of building and exploiting the Semantic Web, including ontology building, ontology modularization, ontology matching, ontology evolution, ontology publication, etc. More recently, he has been working on aspects related to the use of semantic technologies and the Semantic Web for monitoring and managing online personal information. He has also been leading activities around the application of linked data technologies and principles to the Open University, and the education sector in general.
