With the rapid growth of artificial intelligence technology and the deepening integration of education and information technology, the scenario of education has been driven to broaden, which has also prompted the emergence of teaching robot. However, the current research on this pair is mainly focused on the early childhood education and demonstration experimental stage, not applicable to university classrooms, lacking discipline-specific. In order to solve this problem, we designed an artificial intelligence teaching robot system with lecture and Q&A functions, using voice as the overall form of interaction, to teach students the knowledge points while also being able to respond to students' questions, reducing the time cost required to search and filter answers on traditional search engines. For the design of the Q&A function, we use the FAQ Q&A model, build domain-specific knowledge base, and use a combination of inverted indexing and word2vec model to achieve fast question matching and answer extraction. All Q&A situations are recorded by log files to ensure timely checking and updating of the knowledge base. This paper contributes to the educational application of teaching robot in specific subject fields.
Song, Shining and Wan, Yan, "Design And Evaluation Of A Teaching Robot For University Courses" (2022). WHICEB 2022 Proceedings. 88.