Face Expression Classification Based On Behavioral Pattern Recognition Using Deep Learning

1,Department of Computer Systems, Al Nassriyah Technical Institute, Southern Technical University,Thi-Qar, Iraq. 2,3,Department of Information Technology, College of Computer Science and Information Technology, University of Kerbala, karbaka, Iraq. 4,Department of Computer Science, College of Computer Science and Information Technology, University of Kerbala, Karbala, Iraq.

Abstract

In this study, a novel personal identification and engagement monitoring approach based on facial recognition in online examinations is proposed. First, the proposed Back Propagation Convolutional Neural Network (BPCNN) model uses deep transfer learning to perform student identity verification and to unravel emotional cues indicating engagement. The system includes a serious of comprehensive authentication modules to block unauthorized access and guarantee integrity of testing. Furthermore, facial expressions like happiness, anger, sadness, surprise are captured to get the insights into the variations in student engagement. Likewise, the system does not utilize separate facial expression analysis refinements, but indirectly uses emotional cues to improve assessment reliability. The model is shown to have achieved a high accuracy rate of 92% overtaking conventional models against key performance metrics including accuracy, recall, and F1 score. This solution integrates robust facial recognition with indirect emotion based insights for improved security and engagement tracking in e learning environments.