A recommender system for students based on social knowledge and assessment data of competences

Abstract

TEL Recommender systems have been used to improve experiences of students or teachers. Many such systems use information about students, such as interests, preferences, and demographic data. They also use resource metadata and ratings. The authors of this paper think that recommender systems are also valuable when implemented in online or blended courses using competence based assessment since these systems can take advantage of social knowledge about competence development, and students' performance. By using collaborative filtering and knowledge based techniques, it is possible to obtain recommendations from social knowledge and adapt the former to each student's performance. In this paper, the authors propose a system to recommend activities and resources that help students in achieving competence levels throughout an online or blended course. This recommender system takes into consideration experiences previously stored and ranked by former students. In order to offer successful learning advice, this recommender system analyzes the student's current competence levels against similar former students' performances. Functional test results indicate that the proposed technical approach is accurate. Moreover, these results seem to reflect that social knowledge and students' qualifications are sources of valuable recommendations for online and blended courses. © 2014 Springer International Publishing Switzerland.