基于知识模因级联网络的领域知识扩散模式分析

目的/意义]知识经济时代,知识的生产、扩散和消费是社会经济发展的重要推动力。其中,知识扩散是充分发挥知识价值的重要过程,从微观层面理解知识扩散的规律对促进知识利用与创新具有重要意义。方法/过程]以医学信息学领域的科学文献为例,文章基于知识模因的微观层面,采用知识模因识别方法从文献中提取出知识模因来表征知识单元,并基于文献引证网络构建各知识模因的扩散级联网络,计算分析扩散级联网络的基础特征及其特征分布情况,以考察不同知识模因在学科领域内的扩散模式。结果/结论]医学信息学领域内发现了四种典型的知识模因扩散模式:单起源型、多起源-独立型、多起源-迭代型、多起源-融合型。此外,通过对级联网络的各项属性分布特征进行分析发现,网络呈现无标度性,医学信息学领域极少数的知识模因获得了大量的传播资源,领域内的研究集中于被少数知识模因所代表的研究方向上,而领域内各研究方向的生命周期长度则相对差异较小。

Knowledge Diffusion Pattern Analysis Based on Knowledge Meme Cascade Networks

Purpose/significance]In the era of knowledge economy,the production,diffusion and consumption of knowledge are vital driving forces for social and economic development.Among them,knowledge diffusion is an important process to give full play to the value of knowledge.Understanding the pattern of knowledge diffusion from the micro level is of great significance to promote knowledge utilization and innovation.Method/process]Taking the scientific literature of medical informatics as an example,from the microscopic level of knowledge meme,the knowledge meme recognition algorithm is used to extract the knowledge memes from the literature as the knowledge units.And the knowledge meme diffusion cascade networks are constructed based on the citation network,whose basic properties are further investigated to understand the diffusion patterns of different knowledge memes.Result/conclusion]Four typical patterns of knowledge meme diffusion are found in the field of medical informatics:single-origin,multi-origin independent,multi-origin iterative and multi-origin integrated pattern.In addition,through the analysis of the properties’distributions of cascade networks,it is found that cascade networks are scale-free.A small portion of knowledge memes obtain a large amount of resources,and medical informatics focuses on a small number of research directions.And the variance of the length of the life cycle of each research direction is relatively small.