多指标融合的引文网络仿真及领域知识扩散影响因素识别——以知识管理领域为例*

汪 舒 韩 毅

PDF(3421 KB)
图书与情报 ›› 2022, Vol. 42 ›› Issue (05) : 41-50. DOI: 10.11968/tsyqb.1003-6938.2022068
前沿与热点

多指标融合的引文网络仿真及领域知识扩散影响因素识别——以知识管理领域为例*

  • 汪   舒1   韩   毅1,2
作者信息 +

Citation Network Simulation Based on Multi-Indicator Fusion and Influencing Factors Identification to Domain Knowledge Diffusion-Taking Knowledge Management as An Example

  • Wang Shu  Han Yi
Author information +
History +

摘要

在科学交流过程中,知识的价值不仅体现于其内在意义,更体现于引证方式的扩散上,因而以引文网络为基础揭示领域知识扩散机制和影响因素具有重要意义。文章尝试在梳理引文网络形成影响因素的基础上从仿真视角刻画领域知识扩散机制并识别其核心影响因素,探索以文献为载体的知识引用趋势预测新方法。以知识管理领域为例,在中国引文数据库中搜集2010-2020年核心期刊文献,描述真实网络的内容相似性、时间新颖性、位置核心性、作者及团队声誉、期刊影响因子五维指标特征并以此构建引文网络仿真模型,通过蒙特卡洛模拟计算指标最优贡献度。仿真结果表明,真实引文网络的形成与演化受多种因素的影响,其中内容相似性指标的影响贡献度均值最大(26.22%),时间新颖性、作者及团队声誉、期刊影响因子三者的影响相近(约20%),位置核心性的预测贡献性较低(8.11%);多指标融合模型比传统BA模型能更好揭示知识扩散机制。

Abstract

In the process of scientific communication, the value of knowledge lies not only in its intrinsic significance, but also in its diffusion through citation. Therefore, it is significant to reveal the diffusion mechanism and influencing factors in domain knowledge based on citation network. Based on the analysis of the key factors, this paper attempts to depict the mechanism of domain knowledge diffusion from the perspective of simulation, so as to explore new method of knowledge citation prediction based on literature. Taking knowledge management as an example, this paper collects the core journal literatures in this sample from 2010 to 2020 in the China Citation Database, and descripts the statistical characteristics in five dimensions under the context of the real network, i. e., the similarity of literature content, the novelty of literature time, the core of literature position, the reputation of authors and teams, and the impact factor of journals. Based on this, the simulation model is constructed, and the optimal contribution of each indicator is calculated by Monte Carlo simulation. The simulation results show that the formation and evolution of the real citation network are affected by many factors, among which the influence contribution of the similarity of literature content is the highest (Mean=26.22%), the influence of the novelty of literature time, the reputation of authors and teams, and the impact factor of journals is similar (Mean≈20%), and the predictive contribution of the core position of literature is low (Mean=8.11%). The multi-indicator fusion model can better reveal the knowledge diffusion mechanism than the traditional BA model.

关键词

引文网络 / 多指标融合 / 知识扩散 / 系统仿真 / 知识管理

Key words

citation networks / multi-Indicator fusion / knowledge diffusion / system simulation / knowledge management

引用本文

导出引用
汪 舒 韩 毅. 多指标融合的引文网络仿真及领域知识扩散影响因素识别——以知识管理领域为例*. 图书与情报. 2022, 42(05): 41-50 https://doi.org/10.11968/tsyqb.1003-6938.2022068
Wang Shu Han Yi. Citation Network Simulation Based on Multi-Indicator Fusion and Influencing Factors Identification to Domain Knowledge Diffusion-Taking Knowledge Management as An Example. Library & Information. 2022, 42(05): 41-50 https://doi.org/10.11968/tsyqb.1003-6938.2022068
PDF(3421 KB)

602

Accesses

0

Citation

Detail

段落导航
相关文章

/