图书情报工作 ›› 2022, Vol. 66 ›› Issue (4): 103-117.DOI: 10.13266/j.issn.0252-3116.2022.04.011

• 情报研究 • 上一篇    下一篇

面向语义信息分析的多层次技术演化轨迹识别方法研究

马铭1, 王超2, 许海云3, 龚兵营3, 周勇1   

  1. 1. 齐鲁工业大学(山东省科学院)科技发展战略研究所 济南 250014;
    2. 齐鲁工业大学(山东省科学院)情报研究所 济南 250014;
    3. 山东理工大学管理学院 淄博 255000
  • 收稿日期:2021-06-15 修回日期:2021-10-03 出版日期:2022-02-20 发布日期:2022-03-01
  • 通讯作者: 王超,助理研究员,博士,硕士生导师,通信作者,E-mail:wangchao1@sdas.org
  • 作者简介:马铭,硕士研究生;许海云,教授,博士,博士生导师;龚兵营,硕士研究生;周勇,研究员,硕士,硕士生导师。
  • 基金资助:
    本文系山东省科学院青年基金项目"医药领域中突破性技术创新的识别研究"(项目编号:2020QN0012)研究成果之一。

Research on Multi-Level Technology Evolution Trajectory Recognition Method Facing Semantic Information Analysis

Ma Ming1, Wang Chao2, Xu Haiyun3, Gong Bingying3, Zhou Yong1   

  1. 1. Institute of Science and Technology for Development, Qilu University of Technology(Shandong Academy of Sciences), Jinan 250014;
    2. Information Research Institute, Qilu University of Technology(Shandong Academy of Sciences), Jinan 250014;
    3. School of Management, Shandong University of Technology, Zibo 255000
  • Received:2021-06-15 Revised:2021-10-03 Online:2022-02-20 Published:2022-03-01

摘要: [目的/意义] 面向语义信息以层次渐进的方式识别技术演化轨迹,有助于加强对技术细节的理解并提升轨迹识别的准确性。[方法/过程] 首先,提取专利和科技论文的SAO结构,依据语义信息确定研究主题,并利用S曲线分析技术生命周期。其次,借助机器学习算法与社会网络分析指标,分不同周期,通过多层次提取,筛选技术演化轨迹。最后,以造血干细胞领域为实证对象,发现该领域中与遗传病因技术主题相关的专利和科技论文的研究重点存在显著差异,该主题尚未形成统一的演化路径,且有关免疫系统疾病与糖尿病方面的研究是未来潜在的演化趋势。[结果/结论] 所提方法通过客观的数值计算结果,逐步实现复杂技术演化路径的提取与凝练,在揭示技术主要发展历程的同时,能够客观预测技术演化趋势。

关键词: 技术演化, 演化轨迹, 文本分析, 语义信息, 社会网络分析

Abstract: [Purpose/significance] Based on semantic information, identifying the technological evolution trajectory in a hierarchical and gradual manner is helpful to strengthen the understanding of technical details and improve the accuracy of trajectory recognition.[Method/process] Firstly, this paper extracted the SAO structure of patents and scientific papers, determined the research topic based on semantic information, and used the S curve to determine the technology life cycle. Secondly, with the help of the machine learning algorithms and social network analysis indicators, the technology evolution trajectory was extracted and filtered in different cycles and at multiple levels. Taking the field of hematopoietic stem cells as the object of empirical analysis, this study found that there was a significant difference between the research focus of patents and scientific papers related to the subject of genetic etiology in this field. The topic had not yet developed a unified evolutionary path, and the researches on immune system diseases and diabetes were a potential evolutionary trend in the future.[Result/conclusion] The method proposed in this paper gradually realizes the extraction and condensing of complex technological evolution path through objective numerical calculation results. While revealing the main technological development process, method proposed in this paper can objectively predict the technological evolution trend.

Key words: technology evolution, evolution trajectory, text analysis, semantic information, social network analysis

中图分类号: