The delineation of specialties in terms of journals using the dynamic journal set of the SCI
The delineation of specialties in terms of journals using the dynamic journal set of the SCI
复制标题
DOI：
10.1007/bf02016797
复制
发表时间：
2005
期刊：
影响因子：
3.9
通讯作者：
L. Leydesdorff;S. Cozzens
中科院分区：
文献类型：
--
作者：
L. Leydesdorff;S. Cozzens
为了通过使用来自科学引文索引的聚合期刊-期刊引文来将动态期刊集中的期刊归属于专业，有必要用时间序列的视角来补充对这些数据的多变量分析。这就要求在许多可能的聚类参数中选择一个更具分析性的方法。利用多变量分析中的差异，跟踪了科学学科结构的变化。如果这些不确定性事先没有明确定义，就不可能将变化系统地归因于结构、噪音或偏差。本研究报告从概念和方法两方面讨论了必须作出的各种选择。除了期刊之间的层次结构之外，我们还必须假设期刊组（及其质心）之间的分层结构。对于综合制图，定义了作为多维空间中点的密度的表示的“宏轴”的概念。实证结果表明，利用这些方法进行动态期刊映射是可行的。
In order to attribute journals to specialties in a dynamic journal set by using aggregated journal-journal citations derived from theScience Citation Index, it is necessary to complement the multi-variate analysis of this data with a time-serices perspective. This calls for a more analytical approach to the problem of choice among the many possible parameters for clustering. Changes in the disciplinary structure of science are tracked by using thedifferencesamong the multi-variate analyses for the various years. It is impossible to attribute change systematically to structure, noise, or deviance if these uncertainties are not clearly definedex ante. The study discusses the various choices which have to be made, in both conceptual and methodological terms. In addition to hierarchies among journals, one has to assume heterarchy among journal groups (and their centroids). For comprehensive mapping, a concept of “macro-journals” as representations of a density of points in the multi-dimensinoal space is defined. Empirical reslts indicate the feasibility of dynamic journal-journal mapping by using these methods.