A recommendation model for learning objects in a service oriented e-learning environment

This paper proposes a personalized recommendation model to enhance the reusability of the learning objects in a service oriented e-learning environment. The recommendation model is having five main parts. First of all, the model will search for the learning objects that are matched with the user inputted keywords and conditions. Then, the learning objects will be ranked. The model will extract the features of the metadata files of all learning objects in a set while the metadata files of the previous used learning objects into another set. Both sets are used to find out the user´s preference. Then, the model will find the similar users by referring to the User Profile Database. The similar users´ preference will be found out. Both user´s preference and similar users´ preference will be taken into the consideration of the final recommendation score. Ranking of the learning objects is sorted according to the recommendation scores. However, in case if the user does not have any past records in the system (for example new user), the recommendation will be based on the number of overall citation of the learning objects. The contribution of the research is to help users to select and reuse the “best-fit” learning objects. Prototype system will be implemented in future to show the contribution.