Diffusion in heterogeneous networks

We create a program to simulate diffusion in random graphs. Specifically,we create a generalization of bootstrap percolation which incorporates any number of societies and an arbitrary weight matrix W. The code was created from the start to be as general as possible and to easily be modified with further complications. With it we simulate and analyze diffusion in networks consisting of one, two and three distinct societies with different parameters. We study the proportion of adoption as a function of threshold , probability of initial activity and connectivity Pc. We compare several different weight matrices in the two and three society cases to understand the effect of these changes on the diffusion process. The end result is a tool that we propose can be used with the aid of statistics in social or biological contexts to predict behaviors and make conjectures on scenarios not yet observed.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Electrical & Computer Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2015

Author dc:creator

• Dominguez Sagarena, Francisco J.

Contributors dc:contributor

• Baryshnikov, Yuliy

• Bootstrap Percolation
• Computer Simulation
• Diffusion
• Random graphs
• Adjacency matrix
• Heterogeneous
• adoption
• innovations
• Game theory
• threshold
• connectivity

Statement dc:rights

• Copyright 2015 Francisco Dominguez Sagarena

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/87981

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/87981