Identifying and recording user actions to enable automatic online assessment
Lykourgos Petropoulakis, Taha Khan, Linda Steedman · International Journal of Technology Enhanced Learning · 2010
This paper focuses on a system designed to identify and record user actions as they interact with various types of IT training software, and uses this information to provide an automated assessment of these activities in relation to users responding to set tests or exercises. The system has an intelligent inference engine enabling it to either automatically create test templates or analyse submitted tests. The existing prototype differs radically from other similar approaches in that it does not require any specific software platform to operate. Instead, it integrates seamlessly with standard existing business software such as Microsoft Word, PowerPoint or Excel. The present prototype operates on Windows XP and Vista and on the latest versions of Microsoft Office, but it can easily be expanded into other standard software, also used extensively in engineering education such as Matlab and Pspice. The aim of the presented system is to minimise the effort of tutors to set and evaluate tests.