Polytechnic University of Valencia Congress, First International Conference on Higher Education Advances

Font Size: 
Learning Analytics for E-Learning Content Recommendations
Amal Shehan Perera, Sivakumar Tharsan

Last modified: 10-06-2015

Abstract


E-Learning systems have caused a rapid increase to the amount of learning content available on the web. It has become a time consuming and a daunting task for e-learners to find the relevant content that they should study. Existing e-learning technology lacks the automated capability to provide guidance for students to prioritize and engage in the most vital course content. The students who are unable to find out the most suitable resources, for their studies and the assignments, may waste most of their time on browsing and searching. Some of the “good-students” can indirectly act as good guides to other students. Average learners could follow the content adopted by good students in the process of learning. It is possible to capture the behaviour of “good-students” and expose it as a form of automated guiding. For this to work it is important to be able to predict students who are going to be successful at the end of the course based on their performance during the early part of the course. This work demonstrates the use of data mining techniques on e-Learning data to enable “Good-students” to indirectly guide “Average-Students” to find the most relevant content on an e-Learning environment.

DOI: http://dx.doi.org/10.4995/HEAd15.2015.448


Full Text: PDF