Learning engagement: what actions of learners could best predict it?

Research output: Chapter in Book/Report/Conference proceedingChapter (peer-reviewed)peer-review

105 Downloads (Pure)


One important aspect of motivation is engagement. In order to learn, students need to be engaged in the learning activities. However, that does not always happen due to various factors. This paper investigates the possibility to detect the level of engagement of a learner using an e-learning system. More specifically, we are looking for actions that could predict it. Using log files analysis we found that these actions are related to reading pages and taking tests, which are common to every e-Learning system. Several experiments showed that predictions based on attributes related to these two actions are as good as those that include a larger number of actions available in an e-Learning system. A comparison between the attributes found relevant in our research and the attributes used in previous research shows the consistency of our findings. The novelty of our approach is that the focus is on the learning time rather that on evaluation through quizzes-type activities.
Original languageEnglish
Title of host publicationArtificial intelligence in education: building technology rich learning contexts that work
EditorsR. Luckin, K. Koedinger, J. Greer
Place of PublicationWashington
PublisherIOS Press
Number of pages2
ISBN (Print)9781586037642
Publication statusPublished - 2007
EventAIED -
Duration: 1 Jan 2007 → …

Publication series

NameFrontiers in artificial intelligence and applications
PublisherIOS Press


Period1/01/07 → …


Dive into the research topics of 'Learning engagement: what actions of learners could best predict it?'. Together they form a unique fingerprint.

Cite this