Workflow-based assessment of student online activities with topic and dialogue role classification

Abstract

The Pedagogical Assessment Workflow System (PAWS) is a new workflow-based pedagogical assessment framework that enables the efficient and robust integration of diverse datasets for the purposes of student assessment. The paper highlights two particular e-learning workflows supported by PAWS. The first workflow correlates student performance, as measured by project grades, with different dialogue roles, information seeker and information provider, that students take on in project-based discussion forums. The second workflow identifies the distribution of question topics within student discussions. Both workflows employ state of the art natural language processing techniques and machine learning algorithms for dialogue classification tasks. Workflow results were reviewed with a course instructor and feedback regarding the analysis and its fidelity are reported. © 2011 Springer-Verlag Berlin Heidelberg.