Genre Analysis and the Automated Extraction of Arguments from Student Essays

Moreale, Miss E. and Vargas-Vera, Dr M. (2003) Genre Analysis and the Automated Extraction of Arguments from Student Essays. In Christie, Dr J., Eds. Proceedings The Seventh International Computer Assisted Assessment Conference (CAA-2003)., Loughborough University, UK..

This document is not available online.

A full understanding of text is out of reach of current human language technology. However, a shallow Natural Language Processing (NLP) approach can be used to provide automated help in the assessment of essays: our approach uses genre, cue phrases and a set of patterns. Cue phrases, with their associated semantics, are used in conjunction with patterns to identify categories of argumentation partly derived from research in metadiscourse in the academic paper genre. In this paper, we describe an approach for automated extraction of arguments from student essays as a basis for their assessment in a formative as well as a summative sense. We introduce our own essay argumentation schema and show how we arrived at this categorisation. We also introduce "student essay viewer", a tool that allows tutors and students to visualise argumentation in a student essay and may therefore be useful in aiding assessment and providing feedback to students.

Contact the site administrator at: hg@ecs.soton.ac.uk