Smart Learning Object Recommendations based on Time-Dependent Learning Need Models

This paper deals with adaptive learning technologies that fit the individual learner's needs. Thereby, recommender systems play a key role in supporting the user's decision process for an efficient and effective item selection. Activity data have been collected from students using course materials available online. The courses provided access to the course materials via a novel web application. This app also presented learning recommendations to make the content selection more efficient and effective. The paper focuses on the Smart Learning Recommender System which utilizes a novel knowledge-based filtering approach. The algorithm transfers multi-contextual activity data into time-dependent user models. The resulting relevance scores represent the individual user's need to learn specific learning objects at a particular point in time of the course. In comparison to the evaluation results of other educational recommender systems on the same datasets, the introduced approach produces the most precise time-sensitive recommendations.