The Research On Innovation Diffusion Based On Complex Social Network

The evolvement of innovation can divide into two stages:research and development (R&D) and diffusion. R&D is related to the creation of new ideas and subsequent implementation. It is a process for enterprisers to bring an invention into markets. Diffusion refers to the time spread of innovation among social members through certain channels. It is a very important movement process of technology and economy combination, because the social and economic values of innovation can just be realized by diffusion. After entering into 1990s, the enterprises change from conventional R&D mode to exoteric network R&D mode, which forms innovation network. The innovation diffusion among ultimate consumers reckons on consumer network. The changes of consumer network inner structure will influence the innovation diffusion process.The innovation network which comes of R&D cooperation is actually a kind of organization innovation diffusion network. Through cooperation and sharing knowledge, experiences and technology, the enterprises accomplish an R&D project together. So R&D leads to knowledge spillover and further innovation diffusion among enterprises. The two stages of innovation evolvement always go with innovation diffusion, one is the diffusion among enterprises which depends on innovation network; the other is the diffusion among ultimate consumers which depends on consumer network. What this dissertation focuses is a generalized innovation diffusion network, of which consists enterprises innovation network and consumer network.The famous physical scientist Stephen W. Hawking once said the 21st century is a century of complexity. The complex network research is a part of complexity theory. As a powerful tool for studying complexity science and complex system, complex network provide a bran-new angle of view for studying complexity. Some common topology structures that emerge from complex network have not been observed and focused in traditional small size social network analysis. The discovery of these common topology structure characteristics drives the comprehensive and embedded development of social network study on structure and dynamics process, which thus forms a new research direction that is complex social network. Complex social network refers to the social network having complex network characteristics. The theory framework of complex social network method is the microcosmic individuals’ behaviors and mutual influences among them coming forth holistic dynamic behaviors. This framework is just consistent with microcosmic adoption and macroscopic diffusion of innovation.This dissertation is established in the innovation network and consumer network. It studies the evolvement mechanism of innovation network, the relationship between complex consumer network structure and innovation diffusion, the innovation diffusion model based on complex consumer network and enterprise sampling marketing strategy. This study synthetically uses the fruits of complex network, traditional social network and traditional innovation diffusion models. Then we try to search crossed domains among above-mentioned three research aspects and make the innovation diffusion research on a more realistic base. The studies of this dissertation extend application bounds of complex network, and break through existing innovation diffusion research patterns. The results can provide the innovation management of enterprises and innovation policy establishment of government with some theory suggestions. The main research works of this dissertation are as follows:(1) Empirical study on topology structure of China’s investment banks’stock IPO underwriting cooperation network. By building unweighted and weighted investment banks’ stock IPO underwriting cooperation network, we dig out the internal topology structure characteristics of the network. It lays a foundation for further study on establishing innovation network evolvement model.(2) Study on the dynamic evolvement of innovation network. Based on the rational decision-making behaviors of enterprises in the innovation cooperation activities, we establish dynamic evolvement model of innovation network. By using numerical simulation method, we explore the dynamic evolvement laws of innovation network and the relationship between knowledge spillover among enterprises and the network structures. The results demonstrate the innovation networks could self-organized evolve to stable state. The stable innovation networks are small-world networks, but they are not with the highest utilization efficiency.(3) Study on the relationship between consumer network structure and innovation diffusion. Based on the ER stochastic graph model and the idea of WS small world model, we respectively build innovation diffusion network. By simulating the decision-making process of innovation adoption individuals, we explore the influence of the behavior of innovation adoption individuals, consumer network size, structure and character of consumer network on the innovation micro-adoption and macro-diffusion.(4) The innovation diffusion model based on consumer network and empirical study on China’s commercial bank cards. The innovation diffusion model which is based on the consumer network takes into account of both the actual diffusion data and practical consumer network topology structure. It is demonstrated from the empirical study on bank cards of China that the model based on the consumer network fits the actual bank cards diffusion data well. The model reveals different consumer network structures and inner influence mechanism of different bank cards. These results are beneficial to the understanding of financial innovation products’diffusion rules and establishing pertinent marketing strategies.(5) Study on sampling strategies for the diffusion of new products of network externality. We study whether enterprises can learn the optimal sampling targets, taking the scale-free consumer network structure into account, if only aggregate sales data and sampling budget are available. The results reveal the process of enterprises self study and optimize sampling project, including the network position character of optimal sampling targets, the relationship between consumer network topology structure and products diffusion under optimal sampling project, the relationship between optimal sampling and stochastic sampling.Finally, the main contents and results of our research in this dissertation are summarized, as well as future research directions.