English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Student engagement is an important factor in meeting the goals of virtuallearning programs. Automatic measurement of student engagement provides helpfulinformation for instructors to meet learning program objectives andindividualize program delivery. Many existing approaches solve video-basedengagement measurement using the traditional frameworks of binaryclassification (classifying video snippets into engaged or disengaged classes),multi-class classification (classifying video snippets into multiple classescorresponding to different levels of engagement), or regression (estimating acontinuous value corresponding to the level of engagement). However, we observethat while the engagement behaviour is mostly well-defined (e.g., focused, notdistracted), disengagement can be expressed in various ways. In addition, insome cases, the data for disengaged classes may not be sufficient to traingeneralizable binary or multi-class classifiers. To handle this situation, inthis paper, for the first time, we formulate detecting disengagement in virtuallearning as an anomaly detection problem. We design various autoencoders,including temporal convolutional network autoencoder, long-short-term memoryautoencoder, and feedforward autoencoder using different behavioral and affectfeatures for video-based student disengagement detection. The result of ourexperiments on two publicly available student engagement datasets, DAiSEE andEmotiW, shows the superiority of the proposed approach for disengagementdetection as an anomaly compared to binary classifiers for classifying videosinto engaged versus disengaged classes (with an average improvement of 9% onthe area under the curve of the receiver operating characteristic curve and 22%on the area under the curve of the precision-recall curve).

Detecting Disengagement in Virtual Learning as an Anomaly using Temporal Convolutional Network Autoencoder