WDDQA: A Large Language Model-Based Question-answering System in Instruction for the Web Design and Development Course
Year: 2026
Type: conference-paper
Abstract: Large language Models (LLMs) have shown remarkable performance on various natural language processing, including educational question answering. While general-purpose LLMs are capable of addressing certain coding queries, they may not always be optimally suited for specialized educational contexts, such as software technology course at vocational colleges. To address this difficulty, this study introduces a Web Design and Development question-answering System (WDDQA) comprehensively built by ChatGLM3-6B and Langchain-Chatchat. WDDQA assists students in real-time code correction and provide relevant code examples and knowledge. By integrating students’ textbooks with learning data, WDDQA provides explanations that are more accessible and tailored to their comprehension level. Teaching evaluation indicate that after implementing WDDQA in class, the number of student assignment submissions saw an increase of 10%, and the code debugging success rate experienced an increase of 13%. The application of the WDDQA not only enhances student engagement but also improves the quality of code submissions, offering a practical reference for the implementation of AI-supported education.
Authors: Mingpei Peng, Ning Zhang
Institution: Guangzhou Huashang Vocational College
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: hybrid
OpenAlex ID: W7204938907