M4-Week 1-Home

Educational Research Design Module (Week 1)

The following is the homework I carried out on the week immediately after the class that took place on 4th April 2017.

Paper 1 – Introduction
The first paper I used for this week’s homework was written by Professor Vincent Wade (TCD) et al.  I consider it to be a an important paper as it reviewed a number of authoring systems for adaptive learning and personalisation as well as looking at the obstacles to the mainstream adoption of these technologies.

Reference
O’Donnell, E., Lawless, S., Sharp, M., & Wade, V. (2015). A review of personalised e-learning: Towards supporting learner diversity.  Retrieved April 18, 2017, from http://www.tara.tcd.ie/bitstream/handle/2262/73933/odonnell_IJDET%2013%281%29%20article.pdf?sequence=1&isAllowed=y

Definition
The authors define personalisation as the provision of “…each user of a system or the World Wide Web (WWW) with content or an experience which has been tailored to suit their specific needs based on implicit or explicit information about that user…”

Summary
This paper is a review of personalised e-learning, with an impressive reference list of 113 books, articles and web sites spanning 15 years from 1999 to 2013.  It examines some of the technological challenges which software developers may encounter in creating authoring tools for personalised e-learning. It also looks at some of the pedagogical challenges which authors face when creating personalised e-learning activities for students.

Argument that is presented
One of the arguments presented is that learners retain and understand information better by doing something active with it and that the creation of personalised e-learning activities may facilitate active learning.

What I learned from it
That the usability of authoring tools needs to be improved in order to facilitate personalised e-learning.  An ‘adaptive engine’ adapts or dynamically generates the content so that individual users have different learning pathways.

Weakness of one conclusion
The authors refer to an ‘old’ 2005 paper that states that adaptive technologies have only been tested in lab experiments and not by (many) academics.  

How this paper will influence my research proposal
It has focussed my attention on some of the important issues that needs to be addressed before even attempting to design for adaptive learning.

Paper 2 – Introduction
The second paper I used for this week’s homework was written by Professor Peter Brusilovsky (University of Pittsburgh) and his co-author Eva Millán (University of Malaga).  This is considered to be a seminal paper in the field of user modelling.

Reference
Brusilovsky, P., & Millán, E. (2007). User models for adaptive hypermedia and adaptive educational systems. In The adaptive web (pp. 3-53). Springer-Verlag. Retrieved April 18, 2017, from https://pdfs.semanticscholar.org/5cfe/fc79fb172d79c86c17dd2dc1fb6c18786666.pdf

Definition
The user model is a representation of information about an individual user that is essential for an adaptive system to provide the adaptation effect, i.e., to behave differently for different users.

Summary
This paper, which has had 788 academic citations to date, examines the modern approach to user model representation which is called feature-based modelling. This approach attempts to model specific features of individual users such as knowledge and goals.  The older stereotype modelling attempted to cluster all possible users of an adaptive system into several groups called stereotypes.

Argument that is presented
Overlays can be used for user features such as knowledge and interests.  The overlay approach to knowledge modelling uses the domain model and the overlay knowledge model.  The domain model breaks down the body of knowledge about the domain into a set of domain knowledge elements. The overlay knowledge model organises that for each domain model concept, the knowledge component of the user model stores some data that is an estimation of the user knowledge level of this concept.

What I learned from it
I became aware of the five most popular and useful features of the user as an individual are: knowledge, interests, goals, background, individual traits.  These features should be considered when designing for adaptive learning.

How this paper will influence my research proposal
An extension of the overlay knowledge model is a layered model that stores several values to represent user knowledge of each concept.  For example, the system may choose to store separately the levels of user knowledge corresponding to different levels of concept mastery such as Bloom’s Taxonomy of Educational Objectives.  I had previously discussed incorporating Bloom’s Taxonomy in a meeting with Dr. Ioana Ghergulescu, Head of Research and Adaptive Learning with Adaptemy on 13/04/17.

Research Topic
Currently, there is a lot of interest, both academically and commercially, in the field of adaptive learning.  The Horizon 2020 NEWTON project began on 01/03/16 and one its goals is to perform personalisation and adaptation for content, delivery and presentation in order to increase learner quality of experience and to improve learning process.  Three of the fourteen partners are from Ireland:  DCU, NCI and Adaptemy.

My research proposal will be in the area of adaptive learning and personalisation.  Adaptive learning systems are usually divided into separate components or models including domain model (field knowledge), user model (learner profile) and teaching model (pedagogical rules).  I am considering designing and evaluating multi-layered domain and user models and possibly a teaching model, for a well defined part of the Irish second level school curriculum for Mathematics, e.g. functions at Leaving Certificate Ordinary Level.  The final design would be a multilayered network of nodes (knowledge components) and links (connecting these knowledge components) created using graph visualisation/network diagram software such as Graphviz or Microsoft Visio.