Leydesdorff Loet
Amsterdam School of Communication Research (ASCoR), University of Amsterdam, P.O. Box 15793, 1001 NG Amsterdam, The Netherlands.
Scientometrics. 2015;105(3):1441-1451. doi: 10.1007/s11192-015-1639-x. Epub 2015 Jul 22.
In a previous study of patent classifications in nine material technologies for photovoltaic cells, Leydesdorff et al. (Scientometrics 102(1):629-651, 2015) reported cyclical patterns in the longitudinal development of Rao-Stirling diversity. We suggested that these cyclical patterns can be used to indicate technological life-cycles. Upon decomposition, however, the cycles are exclusively due to increases and decreases in the variety of the classifications, and not to disparity or technological distance, measured as (1 - ). A single frequency component can accordingly be shown in the periodogram. Furthermore, the cyclical patterns are associated with the numbers of inventors in the respective technologies. Sometimes increased variety leads to a boost in the number of inventors, but in early phases-when the technology is still under construction-it can also be the other way round. Since the development of the cycles thus seems independent of technological distances among the patents, the visualization in terms of patent maps, can be considered as addressing an analytically different set of research questions.
在之前一项针对光伏电池九种材料技术专利分类的研究中,莱德斯多夫等人(《科学计量学》102(1):629 - 651,2015年)报告了饶 - 斯特林多样性纵向发展中的周期性模式。我们认为这些周期性模式可用于指示技术生命周期。然而,经分解后,这些周期完全是由于分类多样性的增加和减少,而非以(1 - )衡量的差异或技术距离。相应地,在周期图中可显示出单一频率成分。此外,周期性模式与各技术领域的发明人数量相关。有时多样性增加会导致发明人数量增加,但在早期阶段——当技术仍在构建时——情况也可能相反。由于周期的发展似乎独立于专利之间的技术距离,因此专利地图可视化可被视为针对一组分析上不同的研究问题。