Analysis of Behaviors of Agents Affected by Local Interactions Allocated on Scale-Free Networks--Focusing on the Multifractality of Agents' Features--


A - Abstracts of IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences (Japanese Edition)   Vol.J94-A   No.5   pp.323-335
Publication Date: 2011/05/01
Online ISSN: 1881-0195
Print ISSN: 0913-5707
Type of Manuscript: PAPER
local interactions,  scale-free networks,  agents' satisfactions,  multifractality,  

Full Text(in Japanese): PDF(492.4KB)
>>Buy this Article

Many works show the analysis of scale-free properties of networks, but there exit few works related to the generation processes of agents' features. This paper deals with the analysis of behaviors of agents affected by local interactions allocated on scale-free networks where we especially focus on the multifractality of agents' features. At the first model, we assume that there exist a group of agents placed on lattices on a plane who collaborate for the output of products such as software. It is also assumed agents are allowed to move to another sites to maximize their satisfactions. Then, it is shown that the migration patterns of agents bear multifractality. Secondary, we utilize the model where agents react to local interactions by choosing the activities on sites. Then, we see the multifractality in states (activities) of agents. In simulation studies, we show comparative analysis of states of agents based on different network structure of allocations of agents such as scale-free networks and random networks. We also discuss the properties related to real community.