Multi Dimensional Analysis of Learning Experiences over the E-learning Environment for Effective Retrieval of LOs

Abstract

Learner requirements derived through learning experiences plays a major role in any e-learning environment as it has the potential to retrieve the most appropriate Learning Objects (LO) for the learners. With the help of specifications like experience API (xAPI), the e-learning environments these days are capable of recording the learning experiences of the learners both inside and outside the Leaning Management Systems (LMS). These experience statements convey the basic information on the object's utilization to the LMS. However, in a typical e-learning environment with minimal tutor support, such limited information on the learner's experiences may not help the LMSs to determine the dynamically changing needs of its learners. Also, the analysis of collective experiences of similar learners could greatly benefit in determining the learning requirements of a learner. This paper proposes a novel approach towards modeling the learning experience by mapping the learner's profile and the Learning Object Metadata (LOM). The learning experience statements generated on multidimensional perspectives are stored inside the data cube and analyzed using the proposed Cross Dimensional Slicing (CDS) algorithm. The results have highlighted that the learning experiences based LO recommendation has proved to be effective and also reduced the total number of slow learners of the e-learning environment.