Sign in Subscribe
Artificial intelligence

Artificial Intelligence System Evaluates Student Engagement in STEAM Education

Strategic Insights

— 2 min read

Researchers at National Taichung University of Education have developed an AI-based text sentiment analysis system to assess learning engagement in STEAM education. The system integrates speech recognition, natural language processing techniques, keyword analysis, and text sentiment analysis to evaluate the level of learning engagement effectively. The study found that different sentiment dictionaries had a significant impact on the model's accuracy, and the hybrid model proposed in the study outperformed other models in analyzing learners' emotions.

Key Takeaways:

  • The AI system uses sentiment analysis to measure emotional, cognitive, and behavioral involvement in learning, providing insights into student engagement in STEAM education.
  • The system integrates speech recognition, natural language processing techniques, keyword analysis, and text sentiment analysis to evaluate learning engagement levels.
  • The study found that different sentiment dictionaries had a significant impact on the model's accuracy, with the NTUSU sentiment dictionary outperforming other dictionaries in analyzing learners' emotions.
  • The most frequently occurring keywords associated with negative emotions were "problem", "error", "not", and "mistake", indicating that learners experiencing challenges during the learning process are likely to develop negative emotions.
  • The study proposes a hybrid model that utilizes the strengths of SnowNLP and Jieba, indicating a potential for improved sentiment analysis in educational settings.
  • The researchers developed a computational thinking curriculum and study sheets for university students, collecting data on students' participation experiences.
  • The study concludes that teaching materials or courses designed to promote practical, fun, and easy ways of thinking and building logic can help students develop positive emotions and enhance their learning engagement.

Statistics:

  • 1, study participants analyzed in the research
  • 4, different sentiment dictionaries tested in the study
  • 5, models evaluated in the study, including the hybrid model proposed in the research
  • 4304, article reference number for the study (Development of an Artificial Intelligence-Based Text Sentiment Analysis System for Evaluating Learning Engagement Levels in STEAM Education)
  • 15, volume number of the journal Applied Sciences
  • 8, issue number of the journal Applied Sciences
  • 2025, year of publication for the journal article
  • 2025, year of release for the news report

Sources:

  • Development of an Artificial Intelligence-Based Text Sentiment Analysis System for Evaluating Learning Engagement Levels in STEAM Education. Applied Sciences, 2025, 15(8): 4304
  • National Taichung University of Education, Department of Digital Content and Technology
  • MDPI AG, publisher of the journal Applied Sciences
  • Chih-Hung Wu, author, Department of Digital Content and Technology, National Taichung University of Education, Taichung 400, Taiwan
  • Kang-Lin Peng, author, Department of Digital Content and Technology, National Taichung University of Education, Taichung 400, Taiwan

Read more