Inproceedings,

Using Collective Intelligence for Adaptive Navigation in Web Portals

, , and .
Proceedings of the 3rd International Workshop on Adaptation and Evolution in Web Systems Engineering at 8th International Conference on Web Engineering 2008, (Jul 16, 2008)

Abstract

Web Portals offer users a central point of access to companywide information. Initially they focused on presenting the most valuable and widely used information providing quick and efficient information access. But the amount of accessible information has grown tremendously and finding the right information has become more complex and time consuming. We utilize Web 2.0 techniques to address this issue. By incorporating tagging and rating functionality in Portals we can derive users', groups' or entire communities' interests, preferences and skills which allows for reasonable recommendations and adaptations resulting in a more user-tailored Portal. Our aim is to tailor the information presented to the set of information really relevant to a user - especially with respect to a certain context. In particular we analyze users tagging behavior to understand which resources are of higher importance to them than others and provide them with easier access to them. We additionally evaluate the semantic relations between the resources (and tags) being used in order to recommend related resources (or tags) which might be of interest to users in certain situations, too. We finally utilize our knowledge about the semantic interrelation between resources (and tags) to reorder those to better match users' needs. Thus we outline a solution in which we construct a Portal's navigation structure entirely based on the communitys' tagging behavior. Of course, as we do not only analyze each single user's tagging behavior separately, but also the tagging behavior of the entire community, we make use of collective intelligence.

Tags

Users

  • @brusilovsky
  • @aho

Comments and Reviews