Speaker: Thomas Ullman, IET, OU

Abstract:

Promoting reflective thinking is an important educational goal. It is a common educational practice to provide learners with opportunities to express their reflective thoughts in writing. The analysis of such text is mainly a manual task that employs the principles of content analysis.

Considering the amount of text produced by online learning systems, tools that automatically analyse text with regard to reflection would greatly benefit research and practice.

The presentation will report on research that investigated the reliability of machine learning algorithms to detect reflective thinking in text. Surprisingly, the evaluation found that many of the categories that are common in reflective writings can be predicted reliably. The automated method, however, does not achieve the same level of reliability as humans do.

ReflectR – a tool that automatically detects reflection: http://qone.eu/reflectr

Speaker bio:

Dr. Thomas Ullmann is Lecturer at the Institute of Educational Technology at the Open University, UK. He researches especially in the area of Technology Enhanced Learning, with his background in empirical educational science, and computer science. His current focus of research is text analytics for learning. He is especially interested in the automated detection of reflection in writings.

He is the Data Wrangler for the Faculty of Education and Language Studies (FELS) supporting both programmes ‘Childhood, Youth, and Education’, and ‘Languages’ with bespoke Learning Analytics.

His current teaching interests are H818 ‘The network practitioner’ and H817 ‘Openness and innovation in elearning’.

Previously he was working for several EU funded projects, such as the FP7 funded Catalyst project, and the STELLAR Network of Excellence project, as well as the FP6 funded X-Media project.

Profile page: http://qone.eu/ullmann