Collaboratively Shared Information Retrieval Model for e-Learning

Nowadays, the World Wide Web offers public search services by a number of Internet search engine companies e.g. Google [16], Yahoo! [17], etc. They own their internal ranking algorithms, which may be designed for either general-purpose information and/or specific domains. In order to fight for bigger market share, they have developed advanced tools to facilitate the algorithms through the use of Relevance Feedback (RF) e.g. Google’s Toolbar. Experienced by the black-box tests of the RF toolbar, all in all, they can acquire simple and individual RF contribution. As to this point, in this paper, we have proposed a collaboratively shared Information Retrieval (IR) model to complement the conventional IR approach (i.e. objective) with the collaborative user contribution (i.e. subjective). Not only with RF and group relevance judgments, our proposed architecture and mechanisms provide a unified way to handle general purpose textual information (herein, we consider e-Learning related documents) and provide advanced access control features [15] to the overall system.