Clustering as a way to analyze complexity in computer network

This paper proposes an approach to computer network as a complex system that has some level of self-organization where interactions between nodes result in a diverse set of emergent behaviors. The work analyzes artificial computational model of neural networks in order to help us to understand the dynamics of a complex system, and to show their remarkably flexible ability to capture a variety of behaviors of the individuals (neurons) that compose it. The main aim of this work is to make an analogy of the behavior in a neural network and a computer network centered in the organization behavior.