Improving Online Education Using Big Data Technologies
Authors:
- Karim DahdouhKDKarim Dahdouh
- Ahmed DakkakADAhmed Dakkak
- Lahcen OughdirLOLahcen Oughdir
- Abdelali IbrizAIAbdelali Ibriz
Open Access
- 11 March 2020
- book chapter
- Published by IntechOpen
Abstract
In a world in full digital transformation, where new information and communication technologies are constantly evolving, the current challenge of Computing Environments for Human Learning (CEHL) is to search the right way to integrate and harness the power of these technologies. In fact, these environments face many challenges, especially the increased demand for learning, the huge growth in the number of learners, the heterogeneity of available resources as well as the problems related to the complexity of intensive processing and real-time analysis of data produced by e-learning systems, which goes beyond the limits of traditional infrastructures and relational database management systems. This chapter presents a number of solutions dedicated to CEHL around the two big paradigms, namely cloud computing and Big Data. The first part of this work is dedicated to the presentation of an approach to integrate both emerging technologies of the big data ecosystem and on-demand services of the cloud in the e-learning field. It aims to enrich and enhance the quality of e-learning platforms relying on the services provided by the cloud accessible via the internet. It introduces distributed storage and parallel computing of Big Data in order to provide robust solutions to the requirements of intensive processing, predictive analysis, and massive storage of learning data. To do this, a methodology is presented and applied which describes the integration process. In addition, this chapter also addresses the deployment of a distributed e-learning architecture combining several recent tools of the Big Data and based on a strategy of data decentralization and the parallelization of the treatments on a cluster of nodes. Finally, this article aims to develop a Big Data solution for online learning platforms based on LMS Moodle. A course recommendation system has been designed and implemented relying on machine learning techniques, to help the learner select the most relevant learning resources according to their interests through the analysis of learning traces. The realization of this system is done using the learning data collected from the ESTenLigne platform and Spark Framework deployed on Hadoop infrastructure.Keywords
This publication has 11 references indexed in Scilit:
- Smart Courses Recommender System for Online Learning Platform2018
- Big data for online learning systemsEducation and Information Technologies, 2018
- Learning with Big Data Technology: The Future of Education2017
- Big data integration for transition from e-learning to smart learning framework2016
- NIST Big Data Interoperability Framework: Volume 1, Definitions2015
- Using Big Data in the Academic EnvironmentProcedia Economics and Finance, 2015
- Pfp2008
- MapReduceCommunications of the ACM, 2008
- The Google file system2003
- Application-controlled demand paging for out-of-core visualization2002