As chatbots gain prominence across diverse fields, including education, their static
knowledge base poses limitations. This study investigates the utilization of an internet
wizard to enhance the knowledge base of an open-domain question-answering chatbot.
The proposed approach leverages search engines, particularly Google, and their features,
such as feature snippets, knowledge graph, and organic search, by integrating data
science and natural language models. This enables the chatbot to access real-time
and up-to-date answers from web documents, providing dynamic responses to user queries.
A pilot study in higher education assessed the chatbot's mechanism and features, demonstrating
its ability to generate responses across a wide range of educational and non-educational
topics, supported by positive feedback and user satisfaction. The chatbot's dynamic
feature of retrieving related or follow-up questions from search engines significantly
enhances student engagement and facilitates exploration of additional information
beyond the curriculum.
