Scholarlink

<p>The rapid rise of online learning has created both chances and problems. One big problem is that there isn't any real-time emotional feedback, which can make it harder for students to stay interested and learn. This paper suggests HyEmoDetect, a hybrid model for detecting emotions that uses both ''Naïve Bayes and Support Vector Machine (SVM)'' classifiers and improves them through soft voting. The model is trained using the GoEmotions dataset and then used on real-time YouTube Live chat data to accurately figure out how users are feeling. The method focuses on being easy to understand, quick, and able to respond in real time. A visual dashboard shows emotion categories for each class to help teachers change their teaching methods on the fly. In the research we have used metrics like accuracy, precision, recall, and F1-score to judge performance. The results clearly depict that the hybrid model is more accurate than each of the individual classifiers. The system could make real-time feedback systems better in online learning settings. The GoEmotions dataset, a sizable corpus created by Google for emotion classification, is used in this study (Demszky et al. (2020). GoEmotions: A Dataset of Fine-Grained Emotions. arXiv preprint arXiv:2005.00547, 10.48550/arXiv.2005.00547). The GoEmotions dataset includes 58,000 Reddit comments in English that have been manually annotated with a neutral label and 27 different emotion categories. The dataset is available for research use and was downloaded from its official GitHub repository. GoEmotions' emphasis on short-form, real-world textual content that resembles YouTube chat inputs makes it especially appropriate for this study. This work focuses on real-time application, per-class emotional evaluation, and lightweight model architecture intended for inclusion into live online learning environments, unlike previous work.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Real-time emotion detection system using a hybrid machine learning model for enhanced sentiment analysis in online learning environments

  • 期刊论文

  • 发表日期：

    2025年10月25日

查看全文
  • Anand Kumar,
  • Rashmi Agrawal

摘要

The rapid rise of online learning has created both chances and problems. One big problem is that there isn't any real-time emotional feedback, which can make it harder for students to stay interested and learn. This paper suggests HyEmoDetect, a hybrid model for detecting emotions that uses both ''Naïve Bayes and Support Vector Machine (SVM)'' classifiers and improves them through soft voting. The model is trained using the GoEmotions dataset and then used on real-time YouTube Live chat data to accurately figure out how users are feeling. The method focuses on being easy to understand, quick, and able to respond in real time. A visual dashboard shows emotion categories for each class to help teachers change their teaching methods on the fly. In the research we have used metrics like accuracy, precision, recall, and F1-score to judge performance. The results clearly depict that the hybrid model is more accurate than each of the individual classifiers. The system could make real-time feedback systems better in online learning settings. The GoEmotions dataset, a sizable corpus created by Google for emotion classification, is used in this study (Demszky et al. (2020). GoEmotions: A Dataset of Fine-Grained Emotions. arXiv preprint arXiv:2005.00547, 10.48550/arXiv.2005.00547). The GoEmotions dataset includes 58,000 Reddit comments in English that have been manually annotated with a neutral label and 27 different emotion categories. The dataset is available for research use and was downloaded from its official GitHub repository. GoEmotions' emphasis on short-form, real-world textual content that resembles YouTube chat inputs makes it especially appropriate for this study. This work focuses on real-time application, per-class emotional evaluation, and lightweight model architecture intended for inclusion into live online learning environments, unlike previous work.