Group leader prof.dr. P.M.E. de Bra
Paul De Bra studied mathematics and computer science at the University of Antwerp, where he graduated with "greatest distinction". In December 1989 he joined the Computer Science department of the Eindhoven University of Technology, where he heads the Database and Hypermedia Research Group. He performs research on different aspects of hypermedia systems and databases. Currently he is most active in the area of adaptive hypermedia and adaptive Web-based systems. He has been program chair of many important international conferences on hypermedia. He has been steering committee member for EdMedia, and is or has been associate editor and guest editor of several international journals. He is currently vice-president of UM Inc, the non-profit organization that promotes research on and development of User Modeling.
The research
The Databases and Hypermedia chair (DH) concentratse on three aspects of data access and management: (a) the management of semi-structured and semantically linked data, (b) data mining technology for aiding information discovery, (c) user modeling, personalization and adaptive information delivery.
The work on the management of semi-structured (and linked) data concentrates on representing and retrieving information, using semantically meaningful metadata. It ranges from the study of XML, RDF and XMLand RDF query languages (XPath, XQuery, OWL) to concrete strorage, indexing and retrieval technology to effectively and efficiently handle setmi-structured data.
This work has practical applications in Web-Information Systems, where we concentrate on the aspects of navigation and adaptation. We are currently in the process of hiring a new assistant professor to strengthen this research direction.
The work on data mining technology for aiding information discovery combines fundamental research into data structures and algorithms for data mining (to improve the performance of data mining processes) with application-oriented research into specific data mining problems like detecting concept drift and classifying without discriminating. The former has concrete applications in many areas, including detecting trends and fluctuations in industrial processes but also detecting a shift in customer interest in on-line shops and predicting sales in order to optimize (wholesale) store inventory.The latter has applications in e-business where business decisions based on patterns discovered through data mining must obey certain anti-discrimination laws.
For the data mining work we also collaborate with the AIS (Architecture of Information Systems) research group in which the topic of process mining is studied (which is essentially data mining on process data).
The work on personalization and adaptation in information access and delivery concentrates on designing and implementing generic (models and technology for) adaptive information systems. This work has applications in e-culture, e-entertainment, e-learning and e-business.
Personalization is the only way in which humans can continue to use the ever increasing amount of information that is available on-line. Personalization and adaptation not only help the human information consumer but also the providers who can to more targeted information delivery.
Projects in the group can be internal, research-oriented projects, and indutsry-oriented projects, and are defined in depending on the student’s interest.
See the website for more information on the group and its projects,
