As of 30 March, UNESCO estimated that 1.5 billion (87%) of the world’s students had been affected by school/university closures. The bulk of those impacted are in primary and secondary education, with millions also in pre-primary and higher education. Compared to school closures during other historical global crises, the level of education disruption is much greater today.

With the immediate need to move to online learning, student engagement and retention is now a very big challenge. Before the current pandemic, a number of academic studies into learning behaviour and the ability to predict performance of students had already taken place.

What it does

We would like to introduce the e-Learning Insights solution.

e-Learning Insights solution bridges academia and practical use by consolidating data from numerous academic/education administrative systems, such as Blackboard, Canvas, Moodle, etc, to allow academic institutions to not only model student behaviour and predict learning outcomes but also, uniquely, it is able to create profile trait predictions.

By delivering full visibility in a simple to understand dashboard, it is now possible to identify the key learning and engagement attributes of ‘distinction’ students. By using the predictive profiling model, academic institutes can become proactive in their engagement with students as predictive indicator warnings are triggered when there is a change in learning or engagement behaviours by any student, enabling educators to reach out to vulnerable students.

It is recognised that every learning environment and organisation is different. This e-Learning Insights solution provides an intuitive pipeline to enable any e-learning, collaboration or supporting system to provide consolidated data and to recommend and implement the best Machine Learning model for predicting the important outcomes and intervention points for learners.

With e-Learning Insights...

  • Educators and e-learning creators can monitor the engagement and effectiveness of particular e-learning modules within their purview; understand the best time of day/day of week to run virtual classes based on student log on traits; see which e-learning modules and online resources are being accessed and when; and understand if additional support is needed on a particular module or topic thereby improving the students experience.
  • Student welfare teams can be alerted when there is a change in an individual students’ engagement and proactively reach out to them.
  • Faculty management teams can monitor student engagement levels; ensure that key resources are available and utilised; and predict revenue streams including pass/fail/dropout rates.

The e-Learning insights solution includes a unified, complete and predictive view of engagement across an educational establishment — including remote student productivity, changes in learning behaviour and is ideal for educational establishments and examination bodies.

How we built it

The e-Learning Insights has been built as an App on top of the Splunk platform using Splunk Enterprise and the Machine Learning Toolkit.

The prototype uses data from the Open University Learning Analytics Dataset (OULAD). Kuzilek J., Hlosta M., Zdrahal Z. Open University Learning Analytics dataset Sci. Data 4:170171 doi: 10.1038/sdata.2017.171 (2017).

Challenges we ran into

  • Finding suitable test data was the biggest challenge, and we will be looking to partner with many of our existing customers in higher education to develop the concept further with them on a wider range of representative data.
  • Making Machine Learning accessible to non-expert users.

Accomplishments that we're proud of

Making Data Science and Machine Learning accessible and usable for educators - to enable them to focus their efforts on providing effective learning and support.

What we learned

With some further effort to make the content applicable to a wider range of data and working with one of our customers as a partner, we could rapidly develop this into an app that could be used across the education sector.

What's next for e-Learning Insights solution

Technology plays a critical role in keeping essential services functioning and delivering assistance where and when needed, especially at this time, and Splunk is committed to helping in this effort.

We’re ready to collaborate with academic institutions or L&D departments to bring data driven insights to their e-learning use cases at scale. This would include working together to customise the dashboard to ensure this is a reflection of the information that is the most useful for them.

There does appear to be consensus that the provision of e-learning and education, post the COVID pandemic, will change with new opportunities and efficiencies likely to come into play. Clearly, there is room to grow and develop the e-Learning Insights by broadening insights and wider base of users who would find benefit from interpreting and understanding data that’s already available, including:

  • Learners can understand more about their own learning style; engagement with the modules; view their marks and see the prediction of future marks.
  • IT Teams can address performance and real-time availability of e-learning applications; log on and performance data view; fraudulent activity.
  • External examination bodies to verify grades using the student insights data.
  • Government education ministry to understand e-learning effectiveness over time as well as to highlight and share best practice

Try out the prototype here:

  • username: euvsvirus
  • password: EUvsVirusRocks!

This is a functional prototype and may require further scoping and requirements analysis. Therefore, the above is subject to change, agreement and contract.

Built With

Share this project: