The ERC DIADEM Project
How do you go about finding a flat in a foreign city? Use a real-estate aggregator; ask around about real-estate agencies; rummage through the local classifieds … Why isn’t this as easy as searching in Google or shopping at Amazon?
Where the object of publishing is text, the web has become the “great equalizer” and anyone can become a publisher; where the objects are real-word or digital goods, however, this has not manifested. Search engines have not been able to provide comprehensive, automated object search. Though everyone can publish a Web site about a flat for rent, finding that offer in a search engine is nearly impossible.
Why is that? Searching for objects is harder than for text: we search for objects by their attributes—the size of a flat, the number of bathrooms, the distance to the next Opera house. To find an object offered on some Web page, we have to recognize that an object occurs on that page, identify that object’s type and the value of its attributes.
Todays technology requires those tasks to be performed by the publisher. Yet it puts all the power in the hand of large aggregators such as Google: They decide who gets aggregated and how each publisher has to describe its objects. Publishers have to follow the object types defined by aggregators and provide the value of the offered objects; worse, they have to do this for many aggregators.
What
Many computer scientists see the answer for the problem of object search in publishing objects using formal vocabularies. Initiatives such as Linked Open Data and the Semantic Web create such vocabularies for publishers to annotate their published objects.
How
DIADEM is funded primarily through an “Advanced Investigator Grant” of the European Research Council (ERC). The grant finances the majority of the Oxford team as well as visiting collaborators.
Who
The idea of DIADEM is the brainchild of Georg Gottlob, Professor of Computing Science at the University of Oxford and the Vienna University of Technology. His current research deals with web data extraction, database theory, query languages, data exchange, and with graph-theoretic problem decomposition methods that can be used for recognizing large classes of tractable instances of hard problems. He has recently been elected a Fellow of the Royal Society for his fundamental contributions to both artificial intelligence and to database systems.
