USING HYBRID SEMANTIC INFORMATION FILTERING APPROACH IN COMMUNITIES OF PRACTICE OF E-LEARNING

The paper discusses the application of the Information Filtering (IF) approach in Communities of Practice of E-learning (CoPEs). We identify the main characteristics of CoPEs and show how the integration of the IF techniques can be useful in this context as a technology support for members of CoPEs. A personalized recommendation approach is proposed for CoPEs based on the hybrid semantic IF, integrating the content-based filtering, the collaborative filtering and the ontology-based filtering approaches. Three strategies of recommendation have been proposed: (1) a semantic recommendation by specialty; (2) a semantic content-based recommendation by domains of interests; and (3) a semantic collaborative recommendation by domains of interests. We have developed a prototype of a recommendation system called ReCoPESyst, based on the recommendation approach. In order to evaluate our system, we considered a community of teachers from a higher education context. A preliminary tests and experimentation of ReCoPESyst conducted within this community show its advantage and benefit for members.