数学教育学报2026，Vol.35Issue(4)：9-15,7.
面向教学改进的数学课堂评价智能体:框架搭建与效果评价
An Intelligent Agent for Mathematics Classroom Evaluation towards Teaching Improvement:Research on Framework Construction and Effectiveness Evaluation
摘要
Abstract
Classroom teaching evaluation plays a pivotal role in driving the transformation of classroom paradigms,fostering teachers'professional development and elevating teaching quality.However,current classroom teaching evaluation generally relies on experienced teachers,making it difficult to meet the needs of regular teaching improvement.Based on analyzing 149 expert commentaries on teaching cases,we have refined a four-dimensional evaluation framework covering Content Knowledge,Student Learning,Teaching Strategies and Learning Outcomes.Drawing on the mathematics curriculum standards,textbooks and high-quality evaluation corpora as the knowledge base,we have developed a multi-agent-based automated evaluation system for classroom teaching.Benefiting from the construction of an external knowledge base and the multi-agent collaboration mechanism,the intelligent agents can more accurately identify problems in classroom teaching and put forward in-depth,actionable improvement suggestions compared with DeepSeek,a general large language model.These suggestions can serve as a reference for front-line teachers in their teaching improvement and expand the knowledge boundaries of school teaching and research activities.This system is expected to provide low-cost and highly accessible support for classroom teaching evaluation,thereby contributing to the improvement of classroom teaching quality and the balanced development of education.关键词
课堂教学评价框架/智能体搭建/自动化评价Key words
framework for classroom teaching evaluation/LLM agent establishment/automatic evaluation分类
社会科学引用本文复制引用
赵文君,周星宇,曹一鸣,张红..面向教学改进的数学课堂评价智能体:框架搭建与效果评价[J].数学教育学报,2026,35(4):9-15,7.基金项目
国家社会科学基金"十四五"规划2022年度教育学青年课题——人工智能支持下聚焦关键教学行为的课堂教学评价研究(CHA220300) （CHA220300）
