(The) preferentially-attached network grown from a small world작은 세상 네트워크로부터 선호적 연결을 통해 성장하는 네트워크 모델 : 위상특성과 혁신 전파를 중심으로

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 1347
  • Download : 0
Throughout this dissertation, I have developed the network model with separate stages based on a different growing algorithm and have investigated the change of innovation diffusion dynamic by using the two stages evolving network model. In chapter 1, I suggest the growing network model, which use a rewired ring lattice as a seed network and evolve it by using preferential attachment, to describe an evolving process divided into two stages. I then investigate how the structural properties of the modeled network vary with the structure of the initial network. As a result, I show that the topological properties of the initial network remain even after the network growth. However, its vulnerability to preferentially-attached nodes being added is not the same. Moreover, I found out that if a network grown from arbitrary structure, even if nodes are added by preferential attachment, the grown network can be a single scale or broad scale network instead of a scale-free one without the concept of aging or cost of adding links. In chapter 2, I conduct computational simulation experiment of innovation diffusion on the modeled network introduced in chapter 1. As a result, I show that the increase of a portion of preferentially attached nodes decrease the time spent for complete diffusion in most cases. This effect becomes bigger when the initial network is clustered and nodes of network have strong inertia. The results indicate that the efficiency of diffusion can be enhanced as evolution of industry and it can be determined by topological characteristics of early stage and characteristics of nodes. And I compare the influence of topological characteristics of innovator. The results show that the increment of a portion of preferentially attached nodes increases the difference of effectiveness between types of innovators. The results show that diffusing from high-status innovator of core is always an optimal strategy for maximizing innovation diffusion speed. And diffusing from high-status innovator of peripheral shows constantly high effectiveness. In the case of low-status innovator of peripheral and low-status innovator of core, the increment of preferentially attached nodes decreases effectiveness. The results indicate that, as industry evolved, the topological position of innovator on network become more significant for diffusion of innovation.
Advisors
Sung, Kwangjae성광제Yang, Taeyong양태용
Description
한국과학기술원 :기술경영전문대학원,
Publisher
한국과학기술원
Issue Date
2016
Identifier
325007
Language
eng
Description

학위논문(박사) - 한국과학기술원 : 기술경영전문대학원, 2016.2 ,[vi, 91 p. :]

Keywords

Growing network model; Small-world; Preferential attachment; Innovation diffusion; 성장 네트워크 모델; 작은 세상 효과; 선호적 연결; 혁신 전파

URI
http://hdl.handle.net/10203/221091
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=648096&flag=dissertation
Appears in Collection
ITM-Theses_Ph.D.(박사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0