Recommender system for big data in education
Authors:
- 1 August 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Abstract
With the advent of web based e-learning systems, a huge amount of educational data is getting generated. These massive data gave rise to Big data in educational sectors. Currently, big data analytics techniques are being used to analyze these educational data and generate different predictions and recommendations for students, teachers and schools. Recommendation systems are already very helpful in e-commerce, service industry and social networking sites. Recently recommendation systems are proved to be efficient for education sector as well. In this work we are using recommendation system for Big data in education. This work uses collaborative filtering based recommendation techniques to recommend elective courses to students, depending upon their grade points obtained in other subjects. We are using item based recommendation of Mahout machine learning library on top of Hadoop to generate set of recommendations. Similarity Log-likelihood is used to discover patterns among grades and subjects. Root Mean Square Error between actual grade and recommended grade is used to test the recommendation system. The output of this study can be used by schools, colleges or universities to suggest alternative elective courses to students.Keywords
This publication has 10 references indexed in Scilit:
- An Automated Recommender System for Course SelectionInternational Journal of Advanced Computer Science and Applications, 2016
- Educational Data Mining: A Review of the State of the ArtIEEE Transactions on Systems, Man and Cybernetics, Part C (Applications and Reviews), 2010
- Recommender system for predicting student performanceProcedia Computer Science, 2010
- Data Mining2010
- Content-Based Recommendation Systems2007
- Applying Data Mining Techniques to e-Learning Problems2007
- Incorporating contextual information in recommender systems using a multidimensional approachACM Transactions on Information Systems, 2005
- A recommendation algorithm using multi-level association rules2004
- Evaluating collaborative filtering recommender systemsACM Transactions on Information Systems, 2004
- Hybrid Recommender Systems: Survey and ExperimentsUser Modeling and User-Adapted Interaction, 2002