Enhancing learning objects recommendation using multi-criteria recommender systems

To achieve meaningful learning goals, both pedagogues and tutees need frequent supports on how to obtain relevant materials. Recommendation systems have been proved as important tools that assist learners in getting useful learning objects. Nowadays, various recommendation techniques are used to build a system that can find and suggests learning objects to learners. This paper proposed to use a multi-criteria recommendation technique and aggregation function approach for modeling user preferences on learning objects to improve the quality of recommendations given by the existing traditional recommendation systems. The proposed plan is to develop a neural network model and a hybrid of Genetic and Gradient descent algorithms to train the model using real datasets to learn the behavior of the inputs for accurate predictions of learners' preferences.