USP Images

FSTE




Early Warning System

The emergence of e-learning has introduced significant improvements in the way courses are taught and delivered, making this new form of education widely accepted. E-learning systems provide multiple ways of learning (self-paced, collaborative, synchronous & asynchronous, tutorial-based, etc.) and incorporate numerous interactive online activities. Therefore, articulating and identifying user behavior in the e-learning systems can provide new insights into, and improve understanding of how students view and undertake the web-based learning tools and activities. Data accumulated in the system, such as login frequency, participation indexes and completion rates could greatly aid in identifying satisfactory and unsatisfactory performance.

There are two phases of the Early Warning System (EWS)
Phase 1 attempts to improve the quality of online education by introducing innovative mechanisms to motivate and monitor learners progressively according to an acceptable performance threshold. This threshold may be set by the Course Coordinator and may vary from activity to activity and from course to course.

Phase 2 involves the deployment of Machine Learning Techniques (MLTs) to intelligently examine the interaction and activity reports in the Learning Management System (LMS) to diagnose each studentís academic performance. The MLT, precisely Artificial Neural Network (ANN), will be trained on past offering data of the course to derive a predictive map that permits rating studentís ability to pass a course at various points during the semester.


EWS Features

The Early Warning System (EWS) aims to be an intelligent monitoring and feedback system to identify at-risk students during the semester such that corrective actions could be taken before it becomes too late. With many courses venturing into online mode, a self-enthused approach is believed to be more proactive as students can self-monitor their progress and take remedial actions by participating in various online activities.

The systemís functionality:

  • Pops up alerts on upcoming activities and those which are already due.
  • Highlights popular course activities.
  • Measures student engagement with respect to class average.
  • Flags behavioral areas of disengagement (activity participation).
  • Permits the Coordinators to send direct alerts (email/pop-up messages/comments from within Moodle) to under-performing students.
  • Colour codes to quickly identify areas needing intervention.

The module offers the following features for students:

  • Popup alerts on upcoming activities.
  • Overall performance flag
  • Interaction light
  • Login light
  • Completion light
  • Critical activities
  • Popular course activities
  • Engagement/participation graph
  • Login trend graph

For more details see the Student User Manual


Important Dates

EWS Student Workshop
Date:     Friday, 1st August and Monday, 4th August
Time:    11-12pm
Venue:  U8 Lecture Theater

EWS In Refresher and Induction Program
Date:     21st to 25th July
Time:     2-3pm
Venue:  Teaching Lab 1, Japan-Pacific ICT Centre

 


EWS Team

Associate Professor Dr. Bibhya Sharma:
EWS Chair
Assoc. Dean Learning and Teaching
Faculty of Sci., Tech. and Environment
bibhya.sharma(at)usp.ac.fj

Shaveen Singh:
Subject Coordinator
School of Computing,Info & Math Sci
shaveen.singh(at)usp.ac.fj

Vineet Singh:
Research Assistant
Early Warning System
Faculty of Sci., Tech. and Environment
vineet.singh(at)usp.ac.fj


Contact Us

For any queries or information, please contact:
Vineet Singh
Phone: 3231692
Email:   vineet.singh(at)usp.ac.fj


Disclaimer & Copyright l Contact Us l 
© Copyright 2004 - 2017. All Rights Reserved.
Page last updated: Wednesday, July 30, 2014
Faculty of Science, Technology & Environment
The University of the South Pacific
Private Bag, Laucala Campus,
Suva, Fiji.
Tel: (+679) 323 1000
Fax: (+679) 323 1506