Treffer: Dynamic patterns of AI technology diffusion: focusing on time series clustering and patent analysis.

This study examines how the patterns of the technological pervasiveness of AI technology appear over time with patent data from a multi-dimensional perspective. The patent indices of technological pervasiveness here are focused on the process of the recombination and diffusion of technological knowledge via the concepts of generality, originality, complementarity, technology cycle time, and radicalness. For this analysis, the diffusion process of each technology is constructed as time series data, after which dynamic time warping and time series clustering are used for a pattern analysis of the time series data. Also, significant patterns are identified in clusters classified according to time, the diffusion level and the technology type through statistical analytics. As a result, the indices of technology pervasiveness of AI technology as found here show different diffusion patterns depending on time, the diffusion level, and the type of technology. Comprehensively, this study makes useful theoretical and empirical contributions to AI and technology diffusion research given its consideration of the concept of technological pervasiveness as proposed here and the dynamic pattern analysis conducted. [ABSTRACT FROM AUTHOR]