Using Big Data in Collaborative Learning

Using Big Data in Collaborative Learning

Liz Sokolowski (University of West London, UK) and Samia Oussena (School of Computing and Technology, University of West London, UK)
Copyright: © 2016 |Pages: 17
DOI: 10.4018/978-1-5225-0293-7.ch013
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Big data emerged as a dominant trend for predictive analytics in many areas of industry and commerce. The study aimed to explore whether similar trends and benefits have been observed in the area of collaborative learning. The study looked at the domains in which the collaborative learning was undertaken. The results of the review found that the majority of the studies were undertaken in the Computing and Engineering or Social Science domains, primarily at undergraduate level. The results indicate that the data collection focus is on interaction data to describe the process of the collaboration itself, rather than on the end product of the collaboration. The student interaction data came from various sources, but with a notable concentration on data obtained from discussion forums and virtual learning environment logs. The review highlighted some challenges; the noisy nature of this data and the need for manual pre-processing of textual data currently renders much of it unsuitable for automated ‘big data' analytical approaches.
Chapter Preview
Top

Big Data

The term ‘Big Data’ is often described in terms of volume, variety and velocity: Volume: Big Data implies enormous volumes of data being generated primarily by websites, sensors, social media, and so on; Variety refers to the many sources and types of data, both structured and unstructured. Structured data is well defined and can easily be represented as numbers or categories, whereas unstructured can encompass textual information, videos, photos, which creates problems for storing and analysing such data. Velocity refers to the speed with which Big Data is generated - most is produced by machines, rather than by humans and the flow of data is often continuous and massive. Accelerating digitization taking place in all areas of industry and society has meant that Big Data is appearing at an increasing rate in every domain and is available for analysis in ever more creative ways.

This is also true in the context of collaborative learning, where large data sets are potentially available for analysis from students’ interactions with online learning and learning support systems. Learning Management Systems (LMS) such as Moodle and Blackboard record every action a student makes while using the platform, generating a digital footprint of their activity. As well as providing scope for a range of quantitative measures of student activity, these data logs can reveal associations between learners and the structure of networks to which they belong.

In addition, large volumes of textual data can be collected from emails, discussion boards, wikis, phone transcripts, and so forth. As this is primarily unstructured data it needs to be pre-processed and coded in some way prior to analysis.

Complete Chapter List

Search this Book:
Reset