An agent based framework for navigation assistance and information finding in context

Abstract

The main objective of this work is to address the problem of information overload within small groups, driven by similar goals in a way that would enable the delivery of personalised and non-intrusive browsing recommendations and hints as well as aid the users in their information finding activities. The basic idea upon which this work builds is that information gained and created by a user navigating the information space can be used to assist other users in their navigation and information finding activities. The presented model utilises, extends, and combines ideas from open hypermedia with those from Web assistants and recommender systems to achieve its goals. The result of this combination is manifested in the idea of 'linking in context' which this work presents as a novel way of offering Web users recommendations for concepts related to what they are browsing. The integration of the various concepts is facilitated by the use of a multiagent framework.

Creating a flexible and open architecture that can accommodate these goals as well as identifying information finding and recommendation building blocks, is one important dimension of this work. Developing a linking model to embrace context on the user and document level, is another.

More information

Identifiers

Catalogue record

Export record

ASCII CitationAtomBibTeXData Cite XMLDublin CoreDublin CoreEP3 XMLEndNoteHTML CitationHTML CitationHTML ListJSONMETSMODSMPEG-21 DIDLOpenURL ContextObjectOpenURL ContextObject in SpanRDF+N-TriplesRDF+N3RDF+XMLRIOXX2 XMLReferReference ManagerSimple Metadata