Non-Intrusive Identification of Student Attentiveness and Finding Their Correlation with Detectable Facial Emotions
Term of Award
Master of Science in Computer Science (M.S.)
Document Type and Release Option
Thesis (restricted to Georgia Southern)
Copyright Statement / License for Reuse
This work is licensed under a Creative Commons Attribution 4.0 License.
Department of Computer Science
Committee Member 1
Committee Member 2
Teachers use observational cues in the classroom to identify student attentiveness. However, effectiveness of this technique decreases with increasing class size. This work presents an approach for automating these observational cues from the students’ facial expressions and identifying their attentiveness via a neural network machine learning model. Results of the deep learning Convolutional Neural Network model were compared with the range of confidence values obtained from a cloud-based emotion recognition service to perform statistical analysis. The statistical analysis method consisted of correlation analysis which provided the association between the attentiveness level and facial emotions and regression analysis which helped find out the most significant facial emotions. Videos of students were collected during classes and dataset was created for attentive and inattentive students for training the machine learning model. This system can be highly useful for teachers for early identification of inattentive students and taking necessary actions to enhance student learning.
Tabassum, Tasnia, "Non-Intrusive Identification of Student Attentiveness and Finding Their Correlation with Detectable Facial Emotions" (2020). Electronic Theses and Dissertations. 2081.
Research Data and Supplementary Material