pp. 4325-4336
S&M3483 Research Paper of Special Issue https://doi.org/10.18494/SAM4483 Published: December 26, 2023 MCQA: A Responsive Question-answering System for Online Education [PDF] Yi Wang, Jinsheng Deng, Xi Yang, Jianyu Yi, and Zhaohui Ye (Received April 28, 2023; Accepted December 8, 2023) Keywords: QA system, MOOCs, question classification, similarity retrieval, similarity computation, chitchat generation
Massive Open Online Courses (MOOCs) are now considered as representatives of online education. In addition to watching course videos and taking tests, online question & answering (Q&A) also plays an important role during MOOC learning. In this paper, we introduce a question-answering (QA) system called MCQA for MOOCs. The system comprises several modules, including question classification, similarity retrieval, similarity computation, and chitchat generation. On a real MOOC platform, MCQA demonstrates exceptional performance, with experimental results showing a precision rate exceeding 90% and an average duration of a Q&A session of less than 100 ms. Compared with other Chinese-based QA systems, MCQA provides superior open-ended QA capabilities, excelling in performance and covering numerous learning scenarios.
Corresponding author: Yi WangThis work is licensed under a Creative Commons Attribution 4.0 International License. Cite this article Yi Wang, Jinsheng Deng, Xi Yang, Jianyu Yi, and Zhaohui Ye, MCQA: A Responsive Question-answering System for Online Education, Sens. Mater., Vol. 35, No. 12, 2023, p. 4325-4336. |