Title: Objective evaluation of English teaching in colleges and universities based on textual analysis in multi-element perspective
Authors: Xiaohua Chen
Addresses: School of Foreign Languages, Zhanjiang University of Science and Technology, Zhanjiang 524094, China
Abstract: Focusing on the lack of objectivity in college English teaching assessment due to the neglect of multi-modal elements' features in current studies, this paper first designed the cross-modal attention mechanism (CMAM) to learn relevant features among modal elements in the text, and used the gating mechanism to realise the adaptive fusion of multi-modal important information. Then use Transformer model to integrate multi-modal features and modal importance information to realise text sentiment analysis. The analysis results were fused into the evaluation method, and weighted naive Bayes (WNB) algorithm was used to determine the weights of indicators. Finally, a more objective evaluation result is obtained by adjusting the contribution weight of the text to the evaluation result according to students' emotional tendency. The experimental outcome indicates that the F1 of the proposed approach is improved by at least 5.14%, which verifies the effectiveness of the proposed method.
Keywords: English teaching assessment; text analysis; cross-modal attention mechanism; multimodal element; weighted naive Bayes.
DOI: 10.1504/IJRIS.2025.147135
International Journal of Reasoning-based Intelligent Systems, 2025 Vol.17 No.8, pp.11 - 20
Received: 06 May 2025
Accepted: 24 May 2025
Published online: 10 Jul 2025 *