| 基于关键词价值细分的高价值热点主题识别方法研究 |
| |
| 引用本文： | 孙佳佳,李雅静.基于关键词价值细分的高价值热点主题识别方法研究[J].情报学报,2022,41(2):118-129. |
| |
| 作者姓名： | 孙佳佳 李雅静 |
| |
| 作者单位： | 武汉大学信息管理学院,武汉 430072 |
| |
| 摘 要： | 对作者关键词进行价值细分研究，有助于识别学科高价值研究热点主题，帮助研究者们精确把握高价值研究主题和学科研究前沿。本文引入营销领域客户价值细分RFM (recency,frequency,monetary)模型，对各个指标进行动态加权，多次实验后，形成多组关键词价值细分结果；从关键词生命周期的角度，结合医学领域的生存分析方法，使用Kaplan-Meier曲线和Logrank检验验证，识别出最优价值细分结果；依据帕累托原则和聚类算法得到高价值热点主题。数据源选择CSSCI (Chinese Social Sciences Citation Index)收录的图情档领域期刊论文，对1998—2019年的题录数据进行实验。相较于已有的热点主题识别方法，本文的识别结果考虑了关键词的价值属性和分类，较好地识别了高价值热点主题。
|
| 关 键 词： | 价值细分 热点主题 主题识别 |
Recognition of High-Value Hot Topics Based on Value Segmentation of Keywords |
| |
| Sun Jiajia,Li Yajing.Recognition of High-Value Hot Topics Based on Value Segmentation of Keywords[J].Journal of the China Society for Scientific andTechnical Information,2022,41(2):118-129. |
| |
| Authors: | Sun Jiajia Li Yajing |
| |
| Institution: | (School of Information Management,Wuhan University,Wuhan 430072) |
| |
| Abstract: | Research on the value segmentation of authors’ keywords can help identify high-value hot topics, and help researchers accurately grasp high-value topics and research frontiers. By adopting the RFM(recency, frequency, monetary)model for analyzing customer value, this study conducts several experiments by dynamically weighting the indexes of the model, in combination with the survival analysis in the medical field to obtain the best segmentation. According to the Pareto principle and clustering algorithm, the high-value hot topics are identified. The experimental data contain all the records from 1998 to 2019 in the field of Library and Information Science, which are included in the CSSCI database. Compared with the existing recognition methods, the proposed recognition method considers the value attributes and classification of keywords, and better recognizes the high-value hot topics. |
| |
| Keywords: | author keywords value segmentation hot topics topic recognition |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
