图书情报工作 ›› 2013, Vol. 57 ›› Issue (17): 86-94.DOI: 10.7536/j.issn.0252-3116.2013.17.017

• 情报研究 • 上一篇    下一篇

k-clique社区知识创新演化方法研究

白如江1,2, 冷伏海1   

  1. 1. 中国科学院国家科学图书馆;
    2. 山东理工大学图书馆
  • 收稿日期:2013-06-28 修回日期:2013-08-20 出版日期:2013-09-05 发布日期:2013-09-05
  • 作者简介:白如江,中国科学院国家科学图书馆博士研究生,山东理工大学图书馆讲师,E-mail:bairj@mail.las.ac.cn;冷伏海,中国科学院国家科学图书馆研究员。
  • 基金资助:

    本文系山东省自然科学基金项目"大规模学术文献并行处理与语义分类研究" (项目编号:ZR2011GL025)和山东理工大学青年教师发展支持计划研究成果之一。

Knowledge Innovational Evolution Analysis Based on k-clique Community Network

Bai Rujiang1,2, Leng Fuhai1   

  1. 1. Chinese Academy of Sciences, National Science Library, Beijing 100190;
    2. Shandong University of Technology Library, Zibo 255049
  • Received:2013-06-28 Revised:2013-08-20 Online:2013-09-05 Published:2013-09-05

摘要:

提出一种基于k-clique社区的知识创新演化揭示方法。首先,构建科技文献时序关键词共词网络。然后,将共词网络划分为n个最大完整子网络Gs,在Gs中寻找k-clique(2 < k < s)。最后,在给定阈值k的情况下,计算k-clique社区的演化情况,从而揭示知识创新情况。该方法不仅能够有效揭示知识创新演化过程,而且能够通过k-clique社区的关键节点,揭示知识创新过程中的共性知识以及不同知识创新领域的互相影响情况。通过对碳纳米管研究领域2008-2012年SCI数据库论文数据的实验证明,该方法能准确识别出该领域知识创新主要方向,并能准确反映其演化情况。

关键词: 社区网络, 知识创新, 主题演化, 共词网络

Abstract:

This paper proposes a method based on k-clique communities to reveal knowledge innovation and evolution. Firstly, we construct timing keywords co-words network. Then, co-word network is divided into n largest complete sub-network Gs, and looking for K-CLIQUE (2 < k < s) in Gs. Finally, we calculate the evolution of k-CLIQUE communities in the case of a given threshold value, k. This method can effectively reveal the evolution of knowledge innovation and key nodes through k-clique communities overlap. According the structural holes theory, it can reveal the common knowledge, and how it affects each other in the field of innovation. Experiments show that the method can accurately identify the main direction of the field of knowledge innovation, and accurately reflect the evolution of the situation using the SCI database data about carbon nanotube research from 2008 to 2012.

Key words: community network, knowledge innovation, topic evolution, co-words network

中图分类号: