• Sie sind hier:
  • Startseite
  • Studie zu Wissensdiffusion in Netzwerken als Diskussionspapier erschienen

Studie zu Wissensdiffusion in Netzwerken als Diskussionspapier erschienen

Soeben ist die Studie "Simulating knowledge diffusion in four structurally distinct networks – An agent-based simulation model" in der Reihe "Hohenheim Discussion Papers in Business, Economics and Social Science" erschienen.

Die Studie beschäftigt sich mit der Analyse von Wissensdiffusionsprozessen in Netzwerken. Die Ergebnisse der Studie liefern wichtige Impulse für die methodische Umsetzung des VISIBLE Projektes. Erste Ergebnisse der Studie wurden bereits zuvor auf der EMAEE Conference 2015 (European Meeting on Applied Evolutionary Economics) vorgestellt.

Abstract: In our work we adopt a structural perspective and apply an agent-based simulation approach to analyse knowledge diffusion processes in four structurally distinct networks. The aim of this paper is to gain an in-depth understanding of how network characteristics, such as path length, cliquishness and the distribution and asymmetry of degree centrality affect the knowledge distribution properties of the system. Our results show – in line with the results of Cowan and Jonard (2007) – that an asymmetric or skewed degree distribution actually can have a negative impact on a network’s knowledge diffusion performance in case of a barter trade knowledge diffusion process. Their key argument is that stars rapidly acquire so much knowledge that they interrupt the trading process at an early stage, which finally disconnects the network. However, our findings reveal that stars cannot be the sole explanation for negative effects on the diffusion properties of a network. In contrast, interestingly and quite surprisingly, our simulation results led to the conclusion that in particular very small, inadequately embedded agents can be a bottleneck for the efficient diffusion of knowledge throughout the networks.

Keywords: Innovation networks, knowledge diffusion, agent-based simulation, scale-free networks