全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

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

DOI: 10.12677/ae.2024.14122244, PP. 134-139

Keywords: AI辅助课程作业,智能鉴别,高校教育,对比学习
AI-Assisted Course Assignments
, Intelligent Identification, Higher Education, Contrastive Learning

Full-Text   Cite this paper   Add to My Lib

Abstract:

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.

References

[1]  王冰. 生成式AI技术在鞋类配色设计中的应用与分析[J]. 鞋类工艺与设计, 2024, 4(14): 198-200.
[2]  胡正坤, 谭覃. 生成式AI的大视听产业应用和未来展望——以商汤“日日新SenseNova”大模型为例[J]. 视听界, 2024(4): 27-29+38.
[3]  邓元媛, 杨楠, 王子晴. 基于生成式AI的人工智能在建筑设计中的应用探究[J]. 智能建筑与智慧城市, 2024(7): 9-12.
[4]  王宇轩, 徐文浩, 于浩淼, 等. 生成式AI为C语言编程教学带来的挑战和机遇[J]. 计算机教育, 2024(8): 133-141+145.
[5]  徐倩芳. 生成式人工智能在小学信息科技教学中的应用探究[J]. 试题与研究, 2024(22): 121-123.
[6]  Chen, Y., Qian, S., Tang, H., et al. (2024) LongLoRA: Efficient Fine-Tuning of Long-Context Large Language Models. The 12th International Conference on Learning Representations, Vienna Austria, 7-11 May 2014, 1-19.
[7]  北京理工大学. 一种基于写作风格的生成文本来源检测方法[P]. 中国专利, CN202410558437.0. 2024-08-06.
[8]  范志武, 姚金良. 基于深度金字塔卷积神经网络的ChatGPT生成文本检测方法[J]. 数据分析与知识发现, 2024, 8(7): 14-22.
[9]  戎蓉, 杨行, 韩叙, 等. 基于信息熵和GBDT算法的AI生成与人类撰写检测研究[J]. 信息技术与信息化, 2024(7): 186-189.
[10]  李帅彪. 基于TPOC识别与检测人工智能生成内容的研究与探讨[J]. 科技传播, 2024, 16(12): 18-22.
[11]  董腾飞, 杨频, 徐宇, 等. 基于事实和语义一致性的生成文本检测[J]. 四川大学学报(自然科学版), 2023, 60(4): 59-66.
[12]  王一博, 郭鑫, 刘智锋, 等. AI 生成与学者撰写中文论文摘要的检测与差异性比较研究[J]. 情报杂志, 2023, 42(9): 127-134.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133