<![CDATA[The Indonesian Open University currently operates an information service called Hallo UT, which mainly provides academic and administrative information. To improve this service, this study develops a virtual assistant chatbot capable of delivering autonomous 24-hour customer support using a Retrieval-Augmented Generation (RAG) approach. RAG combines information retrieval techniques with large language model capabilities to generate accurate and contextually relevant responses. Data were collected from academic manuals containing frequently asked questions and questionnaires distributed to 76 students. The chatbot was evaluated based on accuracy, response time, and user satisfaction. Results show that the system achieved an average accuracy rate of 92% with an average response time of 5 seconds. In addition, 62% of students responded positively to the chatbot’s functionality. These findings demonstrate the chatbot’s potential to improve student engagement, reduce administrative workload, and enhance the overall learning experience. Future research should involve larger samples, multilingual support, and broader system integration.]]>
Copyrights © 2026
