- About IBCN
- Network Modeling, Design and Evaluation
- Mobile and Wireless Networking
- High Performance Multimedia Processing
- Autonomic Computing and Networking
- Service Engineering
- Content Management and Search
- Data Analysis en Machine Learning
- Information Extraction and Retrieval
- Physical Layer Design
- Application Domains
- Facilities and Tools
Information Extraction and Retrieval
The cluster 'Information Extraction and Retrieval' covers the research domains of text-based search technology and information extraction. The underlying rationale is the need to efficiently search and discover information from large collections of unstructured textual data, be it Web data, news archives, social media data, etc. Due to the 'information overload', this would be impossible without the appropriate technology.
The main research direction is currently towards the automated extraction of (relational) facts from unstructured texts, on various types of content.
In particular, we target social media (e.g., discovering of new places of interest ), news data (e.g., finding relations between people or organizations ), and encyclopedia (e.g., semantic roles ).
On the one hand, the investigation of new applications on realistic content (e.g., continuously crawled from social media) is supported with state-of-the-art machine learning techniques, and on the other hand, we investigate new information extraction techniques (e.g, based on evolutions in the field of deep learning) on internationally recognized test collections.
In the closely related research field of "Information Retrieval" (in the sense of "finding the right content in answer to a user's information need"), our focus is on Federated Web Search  and on evaluation techniques .
Stijn Vandamme, Steven Van Canneyt, Laurent Mertens, Lucas Sterckx, Matthias Feys, Thong Hoang, Cedric De Boom
Selected recent publications
 Van Canneyt S., Schockaert S., Dhoedt B., "Discovering and Characterizing Places of Interest using Flickr and Twitter", International Journal on Semantic Web and Information Systems, 9(3):77-104, 2013.
 Feys M., Sterckx L., Mertens L., Deleu J., Demeester T., Develder C., "UGENT IBCN TAC-KBP 2014 Slot Filling and Cold Start System", Proc. Text Analysis Conference, 2014.
 Vandamme S., De Turck F., "Algorithms for recollection of search terms based on the Wikipedia category structure", Scientific World Journal, 2014.
 Demeester T., Trieschnigg D., Nguyen D., Zhou K., Hiemstra D. “Overview of the TREC 2014 Federated Web Search Track'’, Proc. 23rd Text REtrieval Conference Proceedings (TREC), 2014.
 Demeester T., Aly R., Hiemstra D., Nguyen D., Trieschnigg D., Develder C., "Exploiting User Disagreement for Web Search Evaluation: an Experimental Approach", Proceedings of the 7th ACM international conference on Web search and data mining (WSDM), 2014.