AI辅助课程作业鉴别在高校教育中的应用研究  

Research on the Application of AI-Assisted Course Assignment Identification in Higher Education

作　　者：李娜[1] 王小俊 彭敦陆[1] 

机构地区：[1]上海理工大学光电信息与计算机工程学院,上海

出　　处：《教育进展》2024年第12期134-139,共6页Advances in Education

基　　金：本文得到教育部产学合作协同育人项目(No. 220602510291241)、上海高校青年教师培养资助计划:《数据采集与集成技术》课程思政建设研究与实践(沪教委人[2023] 36号)资助。

摘　　要：为应对AI辅助课程作业给高校教学秩序和学生培养带来的潜在风险,提出一种自动鉴别AI辅助作业的方法。在广泛收集AI生成、AI润色和人类撰写作业的基础上,采用对比学习技术深入挖掘有效区分AI辅助与人类撰写作业的文本特征,并基于这些特征构建一个高效准确的智能鉴别模型。在测试集上的准确率达到92.23%,实现了对不同类别作业的准确鉴别。研究成果不仅为有效的课程作业评估提供准确依据,更为维护公平、良好的教学秩序提供了有力支持。*通讯作者。A method for automatically identifying AI-generated course homework is proposed to address the potential risks that AI-generated homework brings to teaching order and student training. Based on the extensive collection of AI-generated, AI-polished, and human-written homework, contrastive learning technology is used to deeply explore the text features that can effectively distinguish between AI-assisted and human-written homework, and an efficient and accurate intelligent identification model is constructed based on these features. The accuracy on the test set reached 92.23%, achieving accurate identification of different types of homework. The research results not only provide an accurate basis for effective course assignment evaluation but also provide strong support for maintaining fair and good teaching order.

关 键 词：AI辅助课程作业 智能鉴别 高校教育 对比学习 

分 类 号：G63[文化科学—教育学]