Insights and surprises from usage patterns: some benefits of data mining in academic online systems

  • Authors:
  • Owen G. McGrath

  • Affiliations:
  • U.C. Berkeley, Berkeley, CA, USA

  • Venue:
  • Proceedings of the 36th annual ACM SIGUCCS fall conference: moving mountains, blazing trails
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

With the rise of cyber-infrastructure in higher education research and teaching, new challenges surface when it comes to understanding users and usage. How, where, and when user activity gets captured and analyzed in academic online systems is particularly critical in internet-based systems. The flexibility that these open systems allow for in promoting easy integration of different technologies (e.g., applications layer, presentation layer, middleware, and data sources) has repercussions for usage analysis: round the clock access, unseen users, distributed logs, and huge volumes of cryptic data. This paper demonstrates how knowledge discovery solutions - particularly web usage mining methods - have been taken up to address these challenges in one higher education setting involving the Sakai collaboration and learning environment. The goals of this paper include: 1) providing some definitions and explications by example of specific data mining processes as they are actually being used; 2) describing the issues and challenges that motivate the use of data mining and 3) showing how data mining integrates with established project management best practices.