Semantic Mining Based on the Learner´s Preferences

The present research is devoted to development of a technology of the personalized search of learning objects in the Internet, which utilizes the learning object´s and learner´s ontologies, and aggregates the software agents for search of educational materials individually for each learner. The proposed technology manages the personalized search of learning objects for students to support their independent study, for employees to continuously improve their professional skills after graduation, and for teachers to prepare training material. The present paper is focused on the use of learning objects´ and learners´ ontologies developed basing on the OntoClean methodology and IDEF5 method. The software tool OnToIDEF5 has been created for this purpose. These ontologies are devoted to solve the task of personalized search of learning objects on the semantic Web. In order to facilitate this search, we propose a semantic mining approach based on multiple criteria, which assign learner´s preferences using corresponding ontologies. For this purpose we introduced criteria of evaluation of suitability of educational materials to the learner´s profile as well as weighting coefficients for each criterion