Core-network method of empirical research in industrial cluster

Our paper aims to explain how core-network method is valid for the empirical study of cluster network evolution. To be a complex adaptive system, industrial cluster is characterized by its complicated relationship among different kind of organizations within the cluster. However, it is nearly impossible to get the whole-network data through traditional field-survey method not only for the high budget and time limited, but also for the dynamics during cluster evolving process. Therefore, in this paper we try to find the core-network to be an alternative one for empirical study of the cluster. We first put forward the algorithm of core-network and then examine its validity by comparing core-network to whole-network through five statistical indexes including cluster coefficient, average degree, average path length, time efficiency and quality of network, which are borrowed from complex network domain. Finally, the limitation of core-network method and its further research are provided.