Deep Learning Recommendation Model for Education Serious Games for Kids

Abstract

The development in recent years of serious educational games has achieved great success because it has provided positive points for our children by developing themselves, training their characters, and honing their skills. In this context, we need to customize the game related to the child's profile, and we build on the strength of the recommendations system to create personalized recommendations for children during their play.Traditional recommendation systems use a single criteria in the recommendation, while studies have shown that the recommendation to use multicriteria is more accurate. There are many techniques used in the recommendation systems, and the technique that is based on collaborative filtering is the most widely usedOn the other hand, the use of deep learning in recommendation systems began to receive much attention lately. Nevertheless, there is still no attempt to use deep learning in multi-criteria recommendation systems in serious educational games. In order to improve the efficacy of Multi-criteria recommendations system in Educational filed .We interesting to select the most performance functions and techniques for different steps of educational RS.The experiments part showed the results obtains of our comparisons between single criteria recommendation systems and multi criteria RS. In the results section, prove this performance. This is evidence of the successful use of multicriteria accurate technique and recommendation systems with deep learning therefore their dropping in the field of serious games.