Recommender system for ubiquitous learning based on decision tree

In recent years, the fast development of mobile, wireless communication and sensor technologies has provided new possibilities for supporting learning activities. Ubiquitous learning, which is learning that can take place anywhere and anytime, is the best example. In order to provide learners with adequate learning experience, factors such learner's characteristics and context should be considered. Managing the learner context can help delivering the best resource adaptation services. Learning object proposed to the learner is obtained from contexual informations using the decision tree model. On the present paper, a recommender system for ubiquitous learning using context information of the learner and a decision tree model is presented, and k-fold cross validation is used in the experiment for estimating and validating the performance of our recommender system for ubiquitous learning.