Multilingual Sentiment Analysis for Emotion-Aware Feedback in Language Learning Platforms
Abstract
In the growing digital education environment, it is necessary to provide emotionally intelligent ideas to improve learner involvement and motivation. However, most existing language learning sites, particularly in multilingual contexts, overlook the emotional needs of users. This study introduces Muse-Frame (multilingual consciousness and emotional structure for adaptive concept), a new approach that combines emotional analysis with emotional classification, providing real-time, customized ideas. This method involves processing the learner's text in various languages prematurely, utilizing the elegant XLM-Roberta model for emotion and emotional detection, and mapping for adapting to emotional releases. The structure and emotional-awareness results align the detected emotions with the designed ideas, using volume, which improves the user's retention and learning effects. Muse-Frame, evaluated in a combination of multilingual sentiment and learning communication databases, achieves more than 89% accuracy in emotional detection and demonstrates significant improvements in opinion and learner satisfaction. Ultimately, Muse-Frame provides a tangible, intellectual solution for connecting multilingual ideas at modern language learning sites.