iSearch

Satisfying Your Information Needs

The iSearch project gathers our Information Retrieval and Ranking research. We focus on three specific target areas: The World Wide Web, Peer2Peer Networks, and more recently, Personal Computers.

iSearch on the Web. Our main goal on the Web is to propose new solutions for personalizing search. We usually achieve this by means of an improved ranking algorithm built on top of a regular search one. The improvements are either straightforward, or realized through various innovations, which make use of the large scale semantic annotations currently existing on the Web, of the social network information collected for the targeted application, etc. Final output results can then be further enhanced with algorithms for finding related pages, which we are also investigating.

iSearch in a Peer2Peer Environment. We cover most of the research areas related to Peer2Peer networks. The classical search was initiated in the Edutella project, where we were the first ones to bring together the areas of Semantic Web and Peer2Peer, and is now continued here with our Top-K Retrieval research. We are also interested in introducing personalization into the field, both for the search itself and for other applications, such as trust and reputation algorithms. Finally, we deal with continuous querying environments, also known as publish/subscribe systems.

iSearch on the PC Desktop. We are currently opening up towards the very promising area of Personal Computers. The on-going research is about introducing context-based semantics into the desktop search algorithms, but we also investigate other subjects, such as the Social Semantic Desktop, etc. From January 2006, we will closely cooperate with the EU/IST NEPOMUK Integrated Project.

For comments and questions please send an e-mail to chirita [AT] l3s.de