.png){kind=icon}
.png){kind=icon}
Return
Affective Computing-Driven Optimization Methods for Adaptive Foreign Language Learning Systems Research and Empirical Validation
.png){kind=icon}
PRE
AI
DOI：10.1080/10447318.2025.2551048.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This study addresses the challenge of limited emotional interaction in traditional foreign language learning systems by introducing an enhanced approach. The method integrates both multimodal affective computing and reinforcement learning strategies. To thoroughly analyze learner emotions, a three-dimensional emotional mapping model is developed that factors in language anxiety, cognitive load, and learning motivation. Meanwhile, an Emotion-Adaptive Content Generation algorithm (EACG) is designed to create personalized learning materials that dynamically adjust to the learner's emotional state and learning progress. And this study enables the dynamic adaptation of cross-cultural emotional computing models based on an analysis of the emotional characteristics inherent to Chinese, English, and Korean. The experimental findings indicate that multimodal emotion recognition accuracy reaches 89.7%, grammar mastery in the experimental group improves to 85%, peak anxiety levels decrease by 43%, and learning efficiency increases by 37%. These results demonstrate the system's capacity to enhance learning outcomes and emotional adaptation.
Keywords:
Affective computing
foreign language learning
human-computer interaction
multimodal fusion
personalized instruction
Journal
I
IF:
4.9
Papers:
4.3K
Citations:
1.2W