Quantitative Identification of Technological Discontinuities

Abstract

The aim of this paper is to develop and test the metrics to quantitatively identify technological discontinuities in a knowledge network. We first analyzed the various conceptual frameworks for defining such discontinuities and arrived at four metrics. We tested the four metrics: Metric 1 and 2 are the normalized versions of previously existing metrics and Metric 3 and 4 are newly developed from the innovation theories, by using a patent set representative of the magnetic information storage domain. The three representative patents associated with a well-known breakthrough technology in the domain, the giant magneto-resistance spin valve sensor, were selected based on qualitative studies, and the metrics were tested by how well each identifies the selected patents as top-ranked patents. The empirical results show that, first, global citation structure-based metrics clearly provide better performance in the identification of technological discontinuities than local citation count-based metrics which have not been shown as clearly before, second, non-continuous nodes on the major knowledge networks are not at all related to technological discontinuities, and, third, the two global metrics (Metric2: z-score of Persistence and Metric 4: z-score of Persistence times # of converging main paths) successfully identified the three selected patents as top-ranked patents out of over 30 000 patents.

Keywords

• Knowledge discontinuity
• quantitative simulation
• knowledge networks
• patent citation networks