摘要
[目的/意义]定量化、可视化识别基金项目和论文等科技文献主题扩散演化路径,对于揭示科技文献主题之间隐含关联关系辅助研究前沿识别、新兴趋势检测等科技情报分析实践工作具有重要意义。[方法/过程]提出基金项目和论文主题扩散演化路径可视化方法,首先基于LDA模型进行基金项目和论文主题识别,然后利用余弦相似度算法进行基金项目和论文主题关联构建,最后基于Web前端可视化技术设计基金项目和论文主题扩散演化路径可视化方案,揭示基金项目和论文主题的隐含关联关系,辅助进行基金项目和论文主题扩散演化路径分析。[结果/结论]通过对美国近20年纳米农业领域基金项目和论文数据的实证,验证了本文提出方法的准确性和可行性。
[Purpose/Significance]Quantitative and visual identification of the diffusion and evolution path of scientific and technological literature topics such as fund projects and papers is of great significance to reveal the implicit correlation between scientific and technological literature topics,assist in the research of scientific and technological information analysis practice such as frontier identification and emerging trend detection.[Method/Process]The visualization method of diffusion evolution path of fund projects and thesis topics was proposed.Firstly,the fund projects and thesis topics were identified based on LDA model,and then the association between fund projects and thesis topics was constructed by cosine similarity algorithm.Finally,the visualization scheme of diffusion evolution path of fund projects and thesis topics was designed based on Web front-end visualization technology to reveal the implicit association relationship between fund projects and thesis topics,to assist in the analysis of the diffusion evolution path of fund projects and thesis topics.[Result/Conclusion]Through the demonstration of the fund projects and papers in the field of nano agriculture in the United States in recent 20 years,the accuracy and feasibility of the method proposed in this paper are verified.
作者
刘自强
岳丽欣
朱承宁
刘春江
Liu Ziqiang;Yue Lixin;Zhu Chengning;Liu Chunjiang(School of Journalism and Communication,Nanjing Normal University,Nanjing 210023,China;Chengdu Library and Information Center,Chinese Academy of Sciences,Chengdu 610041,China;Library Information and Archives Management,School of Economics and Management,University of Chinese Academy of Sciences,Beijing 100190,China)
出处
《现代情报》
CSSCI
2022年第11期76-86,共11页
Journal of Modern Information
基金
国家自然科学基金项目“基于主题扩散演化滞后的研究前沿识别及趋势预测研究”(项目编号:72104110)
中国科学院文献情报能力建设专项“知识服务主应用系统建设及服务推广”。
关键词
主题识别
关联构建
演化路径
交互可视化
topic identification
association construction
evolution path
interactive visualization