Analysis Of Complex Networks Modeling And Its Application

Complex network has became a hot subject of research on complexity science, and new topics and reserch results are emerging. The research of complex networks modeling has became an important foundation of understanding of the internet networks, social networks, and biological networks.Complex network is divided into non-weighted and weighted complex network. Typical non-weighted complex network contains random network, small world network, BA scale-free network etc, and weighted complex network contains BBV weighted complex network etc.Currently, the research of complex network focuses on weighted complex network rather than non-weighted complex network, and the semantic network of relationships is also a kind of weighted complex network. In this paper, we use the user feedback to build a semantic network that is a weighted complex network, and a weighted network evolution model called UFANE is proposed. UFANE can simulate the behavior of user feedback better, and produce heterogeneous sub-network with different size, density and other characteristics. It has a very strong universal.In this thesis, we also further study the heterogeneous features of the weighted complex network based on user feedback, then propose three new heterogeneous complex network evolution models, compare and evaluate them. By using the search strategies of complex network, we analyze the heterogeneous complex network came from model evolution, and find that there exists "information enter with barrier" in small world network. But in heterogeneous complex networks, the BA scale-free network has obvious advantages in the competition than others. The evolution of heterogeneous complex networks will be towards to BA scale-free network, and the heterogeneous complex network could eventually be homogenized into BA scale-free network.