机器人写作论文新闻效果评价研究——以中国科学报社为例  

Research on Effect Evaluation of Robot-Written Academic Paper News:Taking China Science Daily as an Example

作　　者：张明伟 董雪璠 张其瑶 李濯清 Zhang Mingwei;Dong Xuefan;Zhang Qiyao;Li Zhuoqing(School of Public Affairs,University of Science and Technology of China,Hefei 230026;China Science Daily,Beijing 100190;School of Economics and Management,Beijing University of Technology,Beijing 100124)

机构地区：[1]中国科学技术大学公共事务学院,合肥230026 [2]中国科学报社,北京100190 [3]北京工业大学经济与管理学院,北京100124

出　　处：《图书情报工作》2023年第16期99-110,共12页Library and Information Service

基　　金：中国科学院“十四五”网信专项“数字中科院——智能化科学新闻创作和传播平台建设工程”(项目编号:CASWX2022GC-0305);国家自然科学基金项目“多主体协同视角下中国特色政务信息公开网络舆情治理‘时、度、效’研究”(项目编号:72274010)研究成果之一。

摘　　要：[目的/意义]建立论文新闻评价指标体系,构建综合评价模型,通过实证研究评价机器人写作论文新闻的效果,为机器人更好地写作论文新闻提供参考,促进学术信息生产,推动学术成果传播。[方法/过程]以扎根理论构建指标体系,以层次分析法为指标体系赋权,综合运用模糊综合评价法等构建论文新闻综合评价模型,对“中国科学报”微博和“科学网”微博进行实证研究,对机器人写作论文新闻、人工写作论文新闻和“其他科技新闻”进行对比研究。[结果/结论]研究结果表明,机器人写作论文新闻与人工写作论文新闻和“其他科技新闻”存在差距,但差距不大;论文新闻与“其他科技新闻”相比,存在一定差距;机器人写作论文新闻在时效度方面表现不如预期,在丰富度方面差于人工,在可读度方面表现较差。[Purpose/Significance]This paper establishes an evaluation index system and a comprehensive evaluation model on academic paper news,and conducts empirical studies to evaluate the effectiveness of robot-written academic paper news.It will support the improvement of robot-written academic paper news,the production of academic information,and the dissemination of academic output.[Method/Process]It constructed an index system based on the grounded theory,using analytic hierarchy process to give weight.A comprehensive evaluation model on academic paper news was built by using fuzzy comprehensive evaluation method,among others.Empirical studies were carried out on the microblogs of“China Science Daily”and“ScienceNet.cn”,by comparing robot-written and human-written academic paper news,as well as other science and technology news.[Result/Conclusion]Results indicate that there is a gap between robot-written and human-written academic paper news and other S&T news,but the gap is not too large;there is a mild gap between academic paper news and other S&T news;robot-written paper news does not perform as expected in terms of timeliness,and lags behind in richness comparing to human-written news,and performs badly in readability.

关 键 词：机器人 论文新闻 层次分析法 模糊综合评价 效果评价 

分 类 号：G203[文化科学—传播学]