An architecture for the automated detection of textual indicators of reflection

Ullmann, Thomas Daniel (2011). An architecture for the automated detection of textual indicators of reflection. In: 1st European Workshop on Awareness and Reflection in Learning Networks held in conjunction with the 6th European Conference on Technology Enhanced Learning: Towards Ubiquitous Learning 2011, 21 Sep 2011, Palermo, Italy.



Manual annotation of evidence of reflection expressed in texts is time consuming, especially as fine-grained models of reflection require extensive training of coders, otherwise resulting in low inter-coder reliability. Automated reflection detection provides a solution to this problem. Within this paper, a new basic architecture for detecting evidence of reflection is proposed that allows for automated marking up of written accounts of certain, observable elements of reflection. Furthermore, three promising example annotators of elements of reflection are identified, implemented, and demonstrated: detecting reflective keywords, premise and conclusions of arguments, and questions. It appears that automated detection of reflections bears the potential to support learning with technology at least on three levels: it can foster creating awareness of the reflectivity of own writings, it can help in becoming aware of reflective writings of others, and it can make visible reflective writings of learning networks as a whole.

Viewing alternatives

Download history

Item Actions