Title: Deep learning-based method for detecting student concentration in online learning of English courses

Authors: Xiaohui Zeng

Addresses: School of English Language and Culture, Xi'an Fanyi University, Xi'an, Shaanxi, China

Abstract: To enhance the precision of facial recognition and attentiveness monitoring among students engaged in online English courses, a novel approach leveraging deep learning techniques for assessing student focus during virtual instruction has been developed. Initially, the method employs an image local optimisation preserving photography technique to amalgamate reconstructed facial data into a localised objective function, subsequently segmenting the online learning facial imagery. Subsequently, facial features are extracted using LBP-TOP and fed into a convolutional neural network's SoftMax classifier within the deep learning framework to facilitate facial recognition in the online educational context. Lastly, by consolidating the attentiveness score derived from facial expressions with the score obtained from facial recognition, a comprehensive attentiveness score is computed to gauge student concentration levels. Experimental outcomes indicate that the proposed methodology maintains a facial recognition accuracy exceeding 93%, with the pinnacle of concentration detection accuracy achieving 98.4%.

Keywords: deep learning; English courses; online learning; student concentration detection.

DOI: 10.1504/IJCAT.2024.146136

International Journal of Computer Applications in Technology, 2024 Vol.75 No.2/3/4, pp.114 - 121

Received: 23 Aug 2024
Accepted: 02 Jan 2025
Published online: 07 May 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article