This research aims to develop and evaluate the performance of a Generative AI-powered learning support chatbot for the Thai MOOC platform using Microsoft Copilot Studio. The primary objectives are to enhance online learning experiences and improve support services for both students and educational staff. The chatbot was meticulously designed to provide comprehensive online learning guidance, recommend courses tailored to learners’ profiles and diverse academic interests, and manage an automated issue-reporting system that ensures efficient problem resolution while maintaining high service quality standards. The development used data from over 750 courses on the Thai MOOC platform, employing course metadata (titles, descriptions, objectives) as training material for the chatbot’s recommendation capabilities. Researchers also used system manuals and historical conversation logs to train the chatbot for contextually appropriate guidance and issue resolution. This approach ensured the chatbot effectively addressed learner inquiries. The study involved a trial with 30 volunteer learners who completed post-usage surveys, alongside in-depth interviews with five Thai MOOC staff members. Quantitative and qualitative analyses revealed exceptionally high satisfaction levels among both groups. Learners reported a mean satisfaction score of 4.5 (SD = 0.59), citing the chatbot’s 24/7 availability, instant response capabilities, and precise course recommendations aligned with their academic backgrounds and interests as key strengths. The system demonstrated remarkable accuracy in interpreting user queries, delivering relevant course suggestions, and autonomously logging, categorizing, and escalating issues to relevant teams for swift resolution. All interactions were comprehensively documented for quality assurance purposes. For staff, the chatbot significantly reduced response times and workload by automating routine tasks. However, concerns were raised about long-term maintenance, needing periodic data updates and managing potential scaling challenges. Staff emphasized the importance of establishing clear protocols for updating course information and retraining the chatbot to accommodate platform expansions. This study underscores Generative AI’s potential in developing scalable, efficient online learning support systems like for Thai MOOC. Such tools enrich learner experiences and provide actionable insights for advancing educational technologies. Future iterations could include cross-platform recommendations, enhanced self-repair manuals, and automated email notifications for issue updates, addressing learner feedback.

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

The Integration of Generative AI-Powered Chatbot for Automated Learning Support: Thai MOOC Platform Experience

  • Thapanee Thammetar,
  • Anuchai Theeraroungchaisri,
  • Jintavee Khlaisang,
  • Jira Chonraksuk

摘要

This research aims to develop and evaluate the performance of a Generative AI-powered learning support chatbot for the Thai MOOC platform using Microsoft Copilot Studio. The primary objectives are to enhance online learning experiences and improve support services for both students and educational staff. The chatbot was meticulously designed to provide comprehensive online learning guidance, recommend courses tailored to learners’ profiles and diverse academic interests, and manage an automated issue-reporting system that ensures efficient problem resolution while maintaining high service quality standards. The development used data from over 750 courses on the Thai MOOC platform, employing course metadata (titles, descriptions, objectives) as training material for the chatbot’s recommendation capabilities. Researchers also used system manuals and historical conversation logs to train the chatbot for contextually appropriate guidance and issue resolution. This approach ensured the chatbot effectively addressed learner inquiries. The study involved a trial with 30 volunteer learners who completed post-usage surveys, alongside in-depth interviews with five Thai MOOC staff members. Quantitative and qualitative analyses revealed exceptionally high satisfaction levels among both groups. Learners reported a mean satisfaction score of 4.5 (SD = 0.59), citing the chatbot’s 24/7 availability, instant response capabilities, and precise course recommendations aligned with their academic backgrounds and interests as key strengths. The system demonstrated remarkable accuracy in interpreting user queries, delivering relevant course suggestions, and autonomously logging, categorizing, and escalating issues to relevant teams for swift resolution. All interactions were comprehensively documented for quality assurance purposes. For staff, the chatbot significantly reduced response times and workload by automating routine tasks. However, concerns were raised about long-term maintenance, needing periodic data updates and managing potential scaling challenges. Staff emphasized the importance of establishing clear protocols for updating course information and retraining the chatbot to accommodate platform expansions. This study underscores Generative AI’s potential in developing scalable, efficient online learning support systems like for Thai MOOC. Such tools enrich learner experiences and provide actionable insights for advancing educational technologies. Future iterations could include cross-platform recommendations, enhanced self-repair manuals, and automated email notifications for issue updates, addressing learner feedback.