Creating an academic landscape of sustainability science: an analysis of the citation network
Creating an academic landscape of sustainability science: an analysis of the citation network
复制标题
DOI：
10.1007/s11625-007-0027-8
复制
发表时间：
2007-10-01
影响因子：
6
通讯作者：
Komiyama, Hiroshi
中科院分区：
文献类型：
--
作者：
Kajikawa, Yuya;Ohno, Junko;Komiyama, Hiroshi
可持续性是社会、经济和环境的一个重要概念，每年都会发表数千篇有关该主题的研究论文。随着可持续发展科学成为一个独特的研究领域，明确可持续发展的定义并掌握可持续发展科学的整体结构、现状和未来方向非常重要。本文通过分析学术期刊上发表的论文的引用网络，提供了可持续发展科学的学术图景。拓扑聚类方法用于检测可持续性科学的子领域。结果显示，存在 15 个主要研究集群：农业、渔业、生态经济学、林业（农林业）、林业（热带雨林）、商业、旅游、水、林业（生物多样性）、城市规划、农村社会学、能源、健康、土壤和野生动物。其中农业、渔业、生态经济和林业（农林业）集群占主导地位。从集群中论文的年龄可以看出，能源集群目前正在发展中，尽管其论文数量相对较少。将这些结果与自然语言处理获得的结果进行比较。教育、生物技术、医疗、畜牧业、气候变化、福利和生计集群是通过自然语言处理独特提取的，因为它们是引文网络中跨集群的常见主题。
Sustainability is an important concept for society, economics, and the environment, with thousands of research papers published on the subject annually. As sustainability science becomes a distinctive research field, it is important to define sustainability clearly and grasp the entire structure, current status, and future directions of sustainability science. This paper provides an academic landscape of sustainability science by analyzing the citation network of papers published in academic journals. A topological clustering method is used to detect the sub-domains of sustainability science. Results show the existence of 15 main research clusters: Agriculture, Fisheries, Ecological Economics, Forestry (agroforestry), Forestry (tropical rain forest), Business, Tourism, Water, Forestry (biodiversity), Urban Planning, Rural Sociology, Energy, Health, Soil, and Wildlife. Agriculture, Fisheries, Ecological Economics, and Forestry (agroforestry) clusters are predominant among these. The Energy cluster is currently developing, as indicated by the age of papers in the cluster, although it has a relatively small number of papers. These results are compared with those obtained by natural language processing. Education, Biotechnology, Medical, Livestock, Climate Change, Welfare, and Livelihood clusters are uniquely extracted by natural language processing, because they are common topics across clusters in the citation network.