基于机器学习分类法的AI生成内容与学者撰写内容比较研究  

A Comparative Study of AI-generated Content Based on Machine Learning Taxonomy and Content Written by Scholars

作　　者：王超 潘雪峰 侯辉[2] WANG Chao;PAN Xue-feng;HOU Hui(Liaoning University of Technology Library,Jinzhou 121000,China;Guilin University of Technology Library,Guilin 541004,China)

机构地区：[1]辽宁工业大学图书馆,锦州121000 [2]桂林理工大学图书馆,541004

出　　处：《大学图书情报学刊》2024年第6期9-14,共6页Journal of Academic Library and Information Science

基　　金：2023年度教育部人文社会科学研究规划基金“‘人工智能驱动的科学研究’对科研数据基础设施建设的影响机制研究”(23YJA870002)。

摘　　要：AIGC技术在辅助论文写作中的应用受到广泛关注。文章以信息素养教育领域的论文引言为研究对象,通过对比96篇论文的作者撰写引言与文心大模型生成的引言,利用机器学习分类方法和内容分析技术,深入探讨了两者在语言表达、逻辑结构等方面的异同。对比分析了学者撰写内容与文心一言大模型生成内容的差异性。研究发现大模型生成的引言与学者撰写的引言在语言表达、逻辑结构、信息准确性等方面存在一定相似性,但文心大模型生成的内容在句式变化、深度描述和引用权威性等方面仍有不足。AI工具在辅助论文写作中具有潜力,但学者仍需具备一定的AI素养能力。图书馆作为信息素养教育的重要基地,应加强对读者AI素养的培养,提高其对AI工具的认知水平和伦理意识。The application of AIGC technology in auxiliary paper writing has been widely concerned.This study takes the introduction of papers in the field of“information literacy education”as the research object.By comparing the introduction written by the authors of 96 papers and the introduction generated by ERNIE Bot,this study uses machine learning classification method and content analysis technology to deeply explore the similarities and differences in language expression and logical structure of the two.This paper compares and analyzes the difference between the content written by scholars and the content generated by ERNIE Bot model.It is found that there are certain similarities in language expression,logical structure,information accuracy and other aspects,but the content generated by ERNIE Bot still has shortcomings in sentence pattern change,deep description and authority of quotation.AI tools have the potential in assisting essay writing,but scholars still need to have a certain level of AI literacy.As an important base of information literacy education,libraries should strengthen the cultivation of readersAI literacy and improve their cognitive level and ethical awareness of AI tools.

关 键 词：AIGC技术 信息素养教育 AI素养 机器学习 

分 类 号：G252.7[文化科学—图书馆学]