NORMA eResearch @NCI Library

A Comparison of Traditional Approach and Deep Learning Approach to E-learning Recommender Systems

Pulaparthi, Naga Venkata Satyasainath (2018) A Comparison of Traditional Approach and Deep Learning Approach to E-learning Recommender Systems. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (1MB) | Preview


Nowadays, with rapid growth in learning material and available internet access have made it easy to gain knowledge from online. But, the biggest challenge in e-learning education is how to teach learners effectively. Recommender systems are the most common information filtering systems in several domains to suggest relevant items to users. In the context of an e-learning platform, recommender system is an agent that suggests learners with learning courses based on their interests and previous behavior. The solution to effective teaching problem is preference elicitation that suggests learners based on their desired characteristics. The main objective of this research is to build an accurate recommender system model using a unique deep learning approach, Restricted Boltzmann Machines (RBM) and then compare with successful Matrix Factorization techniques. Our experiment proved that RBM outperformed Singular Value Decomposition (SVD) by 1.6% and Probabilistic Matrix Factorization (PMF) by 5.7%. These results will help data scientists to apply deep learning approach to e-learning recommender platforms.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Q Science > QA Mathematics > Computer software
T Technology > T Technology (General) > Information Technology > Computer software
L Education > LC Special aspects / Types of education > E-Learning
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Caoimhe Ní Mhaicín
Date Deposited: 05 Nov 2018 09:28
Last Modified: 05 Nov 2018 09:28

Actions (login required)

View Item View Item