Premium
Measuring the diffusion of an innovation: A citation analysis
Author(s) -
Zhai Yujia,
Ding Ying,
Wang Fang
Publication year - 2018
Publication title -
journal of the association for information science and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.903
H-Index - 145
eISSN - 2330-1643
pISSN - 2330-1635
DOI - 10.1002/asi.23898
Subject(s) - latent dirichlet allocation , proxy (statistics) , citation analysis , creativity , context (archaeology) , diffusion , computer science , subject (documents) , citation , data science , topic model , psychology , information retrieval , world wide web , social psychology , machine learning , paleontology , physics , biology , thermodynamics
Innovations transform our research traditions and become the driving force to advance individual, group, and social creativity. Meanwhile, interdisciplinary research is increasingly being promoted as a route to advance the complex challenges we face as a society. In this paper, we use Latent Dirichlet Allocation (LDA) citation as a proxy context for the diffusion of an innovation. With an analysis of topic evolution, we divide the diffusion process into five stages: testing and evaluation, implementation, improvement, extending, and fading. Through a correlation analysis of topic and subject, we show the application of LDA in different subjects. We also reveal the cross‐boundary diffusion between different subjects based on the analysis of the interdisciplinary studies. The results show that as LDA is transferred into different areas, the adoption of each subject is relatively adjacent to those with similar research interests. Our findings further support researchers' understanding of the impact formation of innovation.