Recognition and Application of Learner's Cognitive Ability for Adaptive E-learning

In adaptive learning system, the key to promote personalized learning is the learner model. The Elo model has great potential in online learning environment, so based on it, we propose the EELO which is an improved extension of Elo rating system, in view of polychotomously scored items and different granularity evaluations that the Elo rating system do not cover. The performance of the EELO estimating learners' abilities and predicting their future performances are evaluated on two large data set, which demonstrates that the EELO is better-performing. Then, we apply it in a real online learning environment to provide an analysis report for different user roles, which also can be used as the reference for the development of adaptive learning applications in the future.