PERSONALIZED RECOMMENDATION STRATEGIES FOR ELEARNING: AN AHP APPROACH

eLearning has become a key adjunct to both, education in general and the business world; it isbecoming an important tool to allow the flexibility and quality requested by such a kind of learningprocess. One of recent challenges in eLearning industry is personalized learning (PL), aimed onmeeting the needs and aspirations of each individual learner. A PL can be considered as a facility foran individual to access, combine, configure and manage digital resources (knowledge assets andservices) related to their present learning needs and interests. The role of teachers in PL is also enhanced, since they should monitor learners' progress, make dynamic coherence between educational goals and students' achievements, and provide all needed recourses accordingly. The variety of PL systems are already developed, the most attempts of learner personalization are focused on the level of knowledge, background and hyperspace experience, preferences and interests, or even learning styles and achievements. It still does not fully address the issue of inteligent personalized recommendations stimulated by the huge wealth of opportunities for collaboration and communication offered by semantic technologies and intelligent reasoning techniques. In this paper, focusing on the well-known Analytical Hierarchical Process (AHP) method, we propose a framework for addressing different kinds of learners' preferences in PL, integration with historical data and experiences, and making recommendations and personalization accordingly. Firstly, we are focused on making analyses of relevant kinds of preferences defined by both, learners and teachers over learning process in general (including indicators of progress, learning styles, pedagogice approach, etc), learning resources and learners' interests and goals. Also, relevant historical data should be recognized with appropriate retrieving methods and potential web resources if applicable. Finally, semantic structure should be proposed as conceptual framework enabling integration of all gathered data and application of AHP algorithm for processing. The final output of this paper is integrated approach for representing and reasoning over preferences in PL, with effective order decision outcomes in a way that it makes personalized recommendations over available resources,services and dynamic actions.