Teacher-centric educational recommender systems in K12 practice: Usage and evaluation

Abstract

Personalized learning is hard to apply in classroom practice due to the workload of teachers. Recommender systems can help teachers make choices and hence personalize the learning environment. However, few studies have investigated how teachers use recommender systems in their practice. Educational recommender systems also require more investigation into how teachers assess recommendations to provide better systems in the future. Therefore, this research aimed to provide a better insight into the practical application of a teacher-centric educational recommender system, conduct a user-centric evaluation of the recommender system, and present teacher beliefs on the use of educational recommender systems for personalized learning. Our results indicated that teachers similarly used the recommender system by creating new lessons rather than adapting suggested lessons. Their evaluations of the recommender system highlighted the importance of accuracy and utility. Diversity and serendipity were considered to be less important. Teachers perceived recommender systems as a first step to personalized learning and as a tool for simplifying such learning. However, the current system did not affect the quality of personalized learning. Based on these results, we propose to focus more on accuracy and utility as opposed to diversity and serendipity for teacher-centric recommender system evaluation. For personalized learning, we conclude that, according to teachers, recommender systems can lead to more effective personalization by allowing teachers to focus on student needs.