基于PWLR模型的领域新兴趋势识别及其可视化研究  

Identifying and Visualizing Emerging Trends in Domain Based on PWLR Model

作　　者：刘自强 胡正银[2,3] 许海云 方曙[2,3] Liu Ziqiang;Hu Zhengyin;Xu Haiyun;Fang Shu(School of Journalism and Communication,Nanjing Normal University,Nanjing 210097;Chengdu Library and Information Center,Chinese Academy of Sciences,Chengdu 610041;Department of Library Information and Archives Management,School of Economics and Management,University of Chinese Academy of Sciences,Beijing 100190)

机构地区：[1]南京师范大学新闻与传播学院,南京210097 [2]中国科学院成都文献情报中心,成都610041 [3]中国科学院大学经济与管理学院图书情报与档案管理系,北京100190

出　　处：《情报学报》2020年第9期979-988,共10页Journal of the China Society for Scientific and Technical Information

基　　金：中国科学院“十三五”科研信息化专项“面向干细胞领域知识发现的科研信息化应用”(XXH13506-203);国家自然科学基金项目“基于科学—技术主题关联分析的创新演化路径识别方法研究”(71704170)。

摘　　要：探索构建准确、有效的领域新兴趋势分析框架,对于新兴趋势研判、舆情监测等情报工作具有一定的意义。首先,基于N-Gram模型抽取蕴含时间标签的科技文献文本数据中的多元词汇特征Bi-Gram与Tri-Gram;其次,利用分段线性回归(piecewise linear regression,PWLR)模型对Bi-Gram、Tri-Gram进行分段拟合,探测多元词汇在近期时间线上的新兴特征,准确识别有发展潜力的新兴词汇;最后,基于上一步的新兴词汇判别结果利用层次聚类算法识别领域新兴趋势并进行可视化分析。通过对基因编辑领域进行实证研究,识别出基因编辑领域的主要新兴趋势有CRISPR-Cas9技术、基因治疗、动植物基因编辑,验证了本研究提出方法的可行性和有效性。Exploring and constructing an accurate and effective framework for the analysis of emerging trends is of great significance to the information work of emerging trends judgment and research,public opinion monitoring,and so on.First,Bi-Gram and Tri-Gram,which are multi-lexical features,are extracted based on an N-Gram model,and a piecewise linear regression(PWLR)model then is used to fit these features.In addition,the model is used to detect the emerging multi-lexical features on the recent time line as well as to accurately identify new words with potential for development and identify new words based on the previous step.A hierarchical clustering algorithm is used to identify emerging trends in the field and visualize the results.Through empirical research,the main emerging trends in the field of gene editing are identified as CRISPR-Cas9 technology,gene therapy,and animal and plant gene editing,which verifies the feasibility and validity of the method proposed herein.

关 键 词：PWLR模型 新兴趋势 层次聚类 可视化 

分 类 号：G353.1[文化科学—情报学]