Adaptive recommendation of student-created micro-lessons based on learning style and knowledge-level modeling

Intelligent tutoring system (ITS) are increasing in both intelligence and precision over time, as the demand for personalization grows and procedures become more standardized. It is gradually pushing any kind of learning content into the Web because of widespread e-learning. Such materials can be regarded as open educational resources when developed through crowdsourcing methods. This study aims to explore the precision of adaptive learning environments with recommending learning content based on user’s knowledge-level and learning preference. A user-generated micro-lesson project was developed as an alternative to crowdsourcing. Students created their learning objects in the subject: Introduction to the Algorithm. To classify user knowledge level, two machine learning classifier models were evaluated and their accuracy levels have been compared. To assess the effect size of the proposed adaptive system, an experiment was conducted with three instructional groups. The findings suggest satisfactory achievements in control groups. It is concluded that properly addressing learner preferences leads to enhance student behavior and performance in e-learning systems.