Efficient Search in Semistructured Data Spaces
Members
Abstract
An increasing amount of today's data is available in semistructured form, i.e., they provide more structured information than plain text documents, but don't have the regular structure of a relational database. Important examples for such semistructured data are textual data that has been annotated with semantic markup, items like images, movies or bookmarks that have been collaboratively tagged by users in social networks, and knowledge represented in ontologies. Managing and querying large collections of such data requires both effective retrieval models to find relevant results and efficient retrieval algorithms to quickly retrieve these results.
The research done in the group currently focuses on the following topics: (1) efficiently searching in XML data, with a focus on providing answers in (almost) real time, (2) efficiently searching in social networks, considering different relationships of participants in these networks, (3) efficiently searching in large peer-to-peer networks, providing approximate results with limited consumption of network resources, and (4) maintaining persistent archives of highly dynamic document collections, providing efficient and effective search on archived information.