(ProQuest: ... denotes formulae omitted.)
The concepts of regional innovation system (RIS) have been developed into an important framework for evaluating regional performance in the knowledge-based economy from the early 1990s (Cooke, 1992; Cooke, 2001; Cooke & Morgan, 1994). The important elements and mechanisms of a innovation process have been investigated in view of regional-scaled development. Since the early 1990s, the concept of RIS has drawn much attention from policy makers. and it emerged at a time when policy focused toward systemic promotion of localized learning processes in order to establish the competitive advantage of regions (Asheim & Gertler, 2004). RIS approach has received considerable attention as a promising analytical framework for advancing the understanding of the innovation process in the regional economy (Asheim, Coenen, & Svensson-Henning, 2003; Cooke, Boekholt, & Tödtling, 2000; Leydesdorff, 1998).
A lot of attempts have been made to explore ways of mapping knowledge evolution. Author keyword (keywords specified by author), based analysis as a type of co-word analysis has started to play an important role in understanding the dynamics of knowledge development (Hori et al., 2004; Law & Whittaker, 1992; Edquist, 1997). Author keyword analysis is also used to supplement other analytical methods. For example, morphology analysis is a conventional method of forecasting future technology and identifying technology opportunities. Yoon and Park (2004) argued that morphology analysis is subject to limitations because there is no scientific or systematic way of establishing the morphology of technology. Therefore, keyword-based morphology analysis, which is supported by systematic procedures and quantitative data is thereby proposed as a method for conducting the morphology of technology.
Social network analysis based on keywords has been explored as well. Motter et al. (2002) constructed a conceptual network from the entries in a thesaurus dictionary and consider two words connected if they express similar concepts (Motter et al., 2002). He argued that language networks exhibit the small-world property as a result of natural optimization. Hence, these findings are important not only for linguistics, but also for cognitive science. Author keywords, by presenting the most important core concept of the articles' subject, could provide the information about which research trends are of most concern to researchers. The bibliometric method concerning author keywords analysis was developed recently, which uses the author keywords to analyze which trends of research are infrequent (Chiu & Ho, 2007). The technique of author keywords analysis might be a potential method for monitoring development trends or for the evolution of science, as well as for projecting future research directions.
2. RESEARCH QUESTION
The scope of a research field has to be constantly evaluated and redefined due to societal and environmental changes over time. In order to examine the fundamental building blocks of research fields and explore directions toward future research, researchers need to review the literature on a regular basis, and if necessary, modify the scope of research fields in order to obtain stateof- the-art insights.
What are the boundaries and contexts of RIS research? The objective of this study is derived from this research question and aims to analyze the academic literature of RIS research with bibliometric and network analysis to achieve the following purposes: 1) to present an overview of RIS research; and 2) to find the research contexts of RIS. To fulfil the aforementioned objectives, network visualization and analysis software is used to obtain a comprehensive overview of a large amount of literature. The results of this paper visually provide several networks as knowledge maps which define the scopes of RIS research, as well as network properties calculations for quantitatively mapping RIS research. …