Article: Multi-feature fusion model for interactive behaviour recognition in university classrooms using convolutional neural networks and temporal attention mechanisms Journal: International Journal of Reasoning-based Intelligent Systems (IJRIS) 2026 Vol.18 No.14 pp.35 - 48 Abstract: This study proposes a multi-feature fusion model combining convolutional neural networks (CNN) for spatial feature extraction and temporal attention mechanisms for dynamic behaviour modelling in university classrooms. The model was trained and evaluated on a dataset of 5,000 annotated classroom behaviour samples collected from intelligent classrooms across multiple disciplines. Compared to baseline methods such as C3D and CNN-LSTM, the proposed model achieves an F1-score improvement of 5%-9% and increases recognition accuracy to over 90% for key interactive behaviours including hand raising, standing, and note-taking. These results demonstrate the effectiveness of integrating spatial and temporal features for precise classroom behaviour recognition, providing quantitative support for intelligent classroom analysis without relying on subjective evaluation. Inderscience Publishers - linking academia, business and industry through research

Title: Multi-feature fusion model for interactive behaviour recognition in university classrooms using convolutional neural networks and temporal attention mechanisms

Authors: Yuan Luo; Shichao Xu; Maohua Cheng; Dan Ma

Addresses: Guangxi Science & Technology Normal University, Laibin, 546199, China ' Guangxi Science & Technology Normal University, Laibin, 546199, China ' Guangxi Science & Technology Normal University, Laibin, 546199, China ' Guangxi Science & Technology Normal University, Laibin, 546199, China

Abstract: This study proposes a multi-feature fusion model combining convolutional neural networks (CNN) for spatial feature extraction and temporal attention mechanisms for dynamic behaviour modelling in university classrooms. The model was trained and evaluated on a dataset of 5,000 annotated classroom behaviour samples collected from intelligent classrooms across multiple disciplines. Compared to baseline methods such as C3D and CNN-LSTM, the proposed model achieves an F1-score improvement of 5%-9% and increases recognition accuracy to over 90% for key interactive behaviours including hand raising, standing, and note-taking. These results demonstrate the effectiveness of integrating spatial and temporal features for precise classroom behaviour recognition, providing quantitative support for intelligent classroom analysis without relying on subjective evaluation.

Keywords: deep learning; computer vision; university classroom; interactive behaviour recognition; convolutional neural networks; CNN.

DOI: 10.1504/IJRIS.2026.154237

International Journal of Reasoning-based Intelligent Systems, 2026 Vol.18 No.14, pp.35 - 48

Received: 26 Feb 2026
Accepted: 09 Apr 2026
Published online: 17 Jun 2026 *

Title: Multi-feature fusion model for interactive behaviour recognition in university classrooms using convolutional neural networks and temporal attention mechanisms

Keep up-to-date