A Supervised Learning Framework for Learning Management Systems
1. A Supervised Learning framework
for Learning Management Systems
David Monllaó Olivé, Du Q. Huynh, Mark Reynolds, Martin Dougiamas, and Damyon Wiese
School of Computer Science and Software Engineering at The University of Western Australia
Moodle HQ
International Conference on Data Science, E-learning and
Information Systems 2018 (Data'18)
2. 1. Introduction
2. Related disciplines
3. Moodle
4. Framework
5. Framework usage example - Students at risk
6. Future plans
Overview
3. Supervised Learning
“Supervised learning is the machine learning task of learning
a function that maps an input to an output based on
example input-output pairs. It infers a function from labeled
training data consisting of a set of training examples.”
--https://en.wikipedia.org/wiki/Supervised_learning
4. Framework
“abstraction in which software providing generic
functionality can be selectively changed by additional
user-written code, thus providing application-specific
software.”
--https://en.wikipedia.org/wiki/Software_framework
5. Learning Management Systems
“A learning management system (LMS) is a software
application for the administration, documentation, tracking,
reporting and delivery of educational courses or training
programs.”
--Ellis, Ryann K. (2009), Field Guide to Learning
Management, ASTD Learning Circuits
6. Research project purpose
To develop a Supervised Learning framework that
facilitates the creation of predictive models in an LMS.
7. Learning Analytics
"the measurement, collection, analysis and reporting of data
about learners and their contexts, for purposes of
understanding and optimizing learning and the
environments in which it occurs"
--Call for Papers of the 1st International Conference on
Learning Analytics & Knowledge (LAK2011)
8. Educational Data Mining
“an emerging discipline, concerned with developing methods
for exploring the unique and increasingly large-scale data
that come from educational settings and using those
methods to better understand students, and the settings
which they learn in”
--The Educational Data Mining website
9. Moodle
● Open source LMS
● More than 130 million users
worldwide
● My employer :)
● Much more beautiful and usable now
than 2 years ago
https://moodle.net/stats/ - 28th September 2018
11. Framework: Predictive model definition
● Define the target (e.g. late submissions to assignment activities)
● Select, from a list, the indicators (independent variables) that should
predict the target (e.g. number of course accesses, quizzes grades...)
● Set rules to classify data samples in labelled (training) and unlabelled
(receive predictions) (e.g. finished courses / ongoing courses)
● Select when predictions should be generated (e.g. twice a month)
12. Framework: Execution modes
● Testing mode
○ Evaluation of the model predictive power
● Production mode
○ Finished courses data used for training the Supervised Learning
algorithms
○ Insights generated for ongoing courses
13. Framework usage example: Students at risk
● Students at risk of abandoning courses
● Finished courses students that did not log in during the
last quarter of the course are labelled as “at risk”.
● Predictions for ongoing courses are generated after the
1st, the 2nd and the 3rd quarter of the course
15. Future plans
1. Add a UI layer on top of the framework so no coding is
required for simple predictive models
2. Add more predictive models to Moodle core
3. Improve the Machine Learning backend layer