Modeling Small-World Trust Networks

Constructing a trust network can provide a strong foundation for realizing security in ubiquitous environments. This approach is inspired by social networks, which have two typical characteristics: small-world and scale-free properties. Small-world properties consist of high clustering and short average path length. It often means that we can easily find a short path between any two agents in such networks, which is an expected property of trust networks. Another typical feather of social networks is scale-free link distribution. The most notable characteristic of scale-free networks is power law distribution, which means there are some nodes with a degree that greatly exceeds the average. Several models have been proposed to capture small-world and scale-free properties respectively, but each kind of models often lacks the property of the other kind of models. In this paper, based on the existing research results of small-world and scale-free models, we research how to model a trust network with both small-world and scale-free properties. Analyses demonstrate that the approach in this paper is efficient and effective.