基于作者特征和期刊特征的学术论文被引频次预测模型构建与分析  

Construction and Analysis of Forecasting Model for Journal Paper Cited Frequency Based on Author and Journal Characteristics

作　　者：程子轩 张向先[1] 郭顺利[3] CHENG Zi-xuan;ZHANG Xiang—xian;GUO Shun-li(School of Management,Jilin University,Changchun 310022,China;Changchun University,Changchun 310022,China;Media College,Qufu Normal University,Rizhao 276826,China)

机构地区：[1]吉林大学管理学院,吉林长春310022 [2]长春大学,吉林长春310022 [3]曲阜师范大学传媒学院,山东日照276826

出　　处：《情报科学》2021年第3期179-184,192,共7页Information Science

基　　金：吉林省高教科研一般课题“基于SPOC的混合式教学在物联网工程仿真技术课程中的应用研究”(JGJX2019D416)。

摘　　要：【目的/意义】预测学术论文的被引频次不仅能够帮助作者提前了解个人研究成果未来的应用价值,还能够帮助期刊编辑人员预估期刊的影响力。【方法/过程】通过分析论文作者特征、论文所在期刊特征与论文被引频次的相关性,发现影响论文被引频次的重要因素;利用逐步回归分析建立用于预测论文被引频次的数学模型,用于预测论文未来几年内的被引频次。【结果/结论】实验发现与学术论文被引频次呈显著相关的10个因素;发现可用期刊即年指标和作者之前发表论文的篇均被引频次的线性组合来预测学术论文未来的被引频次。【创新/局限】分析预测结果,发现可通过以下途径提高预测模型的准确度:将某学科的所有期刊限定到学科内某一特定期刊、将作者之前发表论文的篇均被引频次限定为以第一作者身份发表论文的篇均被引频次、将论文的研究内容纳入预测模型等。【Purpose/significance】In order to help the author understand the future value of personal research in advance and help journal editors predict the impact of journals, we propose the method for predicting citation frequency of academic papers.【Method/process】In this paper, we find that the important factors affecting the journal paper cited frequency by analyzing the correlation between the author characteristics and papers cited, periodical characteristics and papers cited;and establish the mathematical model for forecasting the journal paper cited frequency by stepwise regression analysis.【Result/conclusion】This experiment found 10 factors which significantly correlated with the cited frequency of journal papers;the future cited frequency of journal papers can be predicted using the linear combination about the periodical immediacy index and the average cited frequency per paper.【Innovation/limitation】The prediction model still has errors even it has a certain coincide with true value. Finally, we provide the effective way to improve the accuracy of prediction by analyzing the causes of error.

关 键 词：作者特征 期刊特征 预测模型 相关分析 逐步回归分析 

分 类 号：G250.2[文化科学—图书馆学]