Ed Wakelam

Researching the achievement of pedagogy led step change in the application of artificial intelligence techniques to learning/training systems.  Prior to this I had a 39 year career spanning operations, programme and strategic management, software and systems development through to board level responsibility.  Delivering high quality asynchronous training within global organisations is an increasingly important component of corporate business as is the growing trend of e-Learning in Education.

I believe that there is considerable potential to make significant steps forward in the application of Artificial Intelligence (AI) to learning systems. A variety of AI techniques can be applied in real-time to analyse learner behaviour, tailor learning components to learner abilities and knowledge, and to exploit the very large quantities of subject and student data available in both the education and commercial sectors. The development of learning systems in conjunction with trainers, teachers and subject matter experts will provide benefits to institutions across the board, from career/vocational development, re-validation and re-training through to higher education and school.

I believe that this potential has been there for some time, and while my research to date has found a variety of work discussing and modelling how individual AI techniques can be applied to different aspects of learning systems and student achievement, I have identified very few examples of comprehensive learning systems that exploit AI techniques.

My research has indicated that bridging the gap between brilliant new techniques in AI and Machine Learning (ML) such as neural networks, data mining, fuzzy logic and ant colony optimisation and the essential pedagogy (the theory and practice of education) proves to be a significant challenge.

However, I believe that the opportunity to make step change progress is now much stronger with the convergence of several of the required components for success increasingly in place.  These are:

• The availability of appropriate learning platforms, with almost all learners having computing devices both inside and outside of the learning setting.
• The increasing quantity and quality of the data (subject and analytics) available to the analytical learning systems using AI.
• The technology (hardware and supporting software) is now powerful enough to handle and exploit the quantity and complexity of data and algorithms necessary for success.
• Institutions are putting more emphasis into this area – exploiting e-learning opportunities and looking for efficiency gains.
• Learners are increasingly interested in learning and developing their knowledge on-line at least in parallel with the traditional classroom/campus model.

As a result, I believe that the deployment of AI and ML techniques in Technology Enhanced Learning (TEL) is poised for accelerated growth and adoption.

In particular, I want to explore how AI and ML techniques can be applied to the development of adaptive learning systems, including the classification and representation of subject matter knowledge. By the latter I am referring to the organisation of the subject knowledge, the rules and the processes which connect them into a logical structure that:

• Is comprehensive and efficient for the learning system, as well as for the creation, validation and future manipulation by the subject matter expert (SME).
• Is capable of incorporating all the relevant interconnections between the information in a similar way to the way our own brains do so.
• Allows the learning system itself to automatically self-organise and search for further connections and rules.

I believe that we have yet to see much of the excellent progress in AI and ML techniques being deployed in the development of learning systems. My research question is: “How can leading edge artificial intelligence techniques be used to improve technology enhanced learning?”