Directed Weighted Complex Network Survivability Measure

This paper summarizes the measure algorithms of invulnerability based on graph theory and statistical physics from the basic theory of complex network. Puts forward the concept which we call loop coefficient of node to reflect the paths one node may choose and divides from the harmony shortest distance of the directed and weighted network to reflect the strength of the links between the nodes by anglicizing the structure of the directed and weighted complex networks and the effect of the important parameters such as out-degree, in-degree and the weight of edge.At last, this paper proposes an algorithm which is used to measure the invulnerability of directed and weighted complex network—IMADW measure algorithm and deal with the results normalize. This paper divides research process of the invulnerability into three phases such as preparatory phase, implementation phase, and validation phase, the specific content arrangement as follows:(1) Preparation phase. The mainly work of this phase is to accumulate the theoretical, including basic concepts of complex networks, research status of the invulnerability. In order to understand the structure of the directed and weighted complex network fully, this paper chooses the more popular research model (modified of the BBV model) of the complex network currently to perform and analyze many important control parameters of the model and gives some ideas about the weight of the network.(2) Implementation phase. The mainly work of this phase is to give the algorithm design idea, including the measurement, the evaluation index, the algorithm analysis of the basic factors and the design idea of IMADW measure algorithm.(3) Validation phase. The mainly work of this phase is to verify the validity of the algorithm and build a simulation platform to verify the algorithm from multi-angle. The experiment need to generate a directed and weighted network and analyze the basic parameters of the network. Meanwhile, the paper also analyzes and evaluates the experimental data sets——Facebook-like e-mail and gives the effectiveness and limitations of the IMADW measure algorithm.