A Hierarchical Random Graph Efficient Sampling Algorithm Based on Improved MCMC Algorithm

A hierarchical random graph (HRG) model combined with a maximum likelihood approach and Markov Chain Monte Carlo algorithm can not only be used to quantitatively describe the organization of many real networks, but also predict missing connections in partly known networks high accuracy. However, computational cost is very large when graphs are sampled by (MCMC), so that graphs, which characteristics network structure, cannot found reasonable time range. This seriously limits practicability model. In order overcome this defect, an improved MCMC called two-state transitions (TST-MCMC) for efficiently sampling proposed paper. On chain composed all possible TST-MCMC generate two candidate state variables during transition introduce competition mechanism filter out worse variables. addition, detailed balance ensured using Metropolis–Hastings rule. By method, convergence speed improved, interval narrowed as well. Three example employed verify performance algorithm. Experimental results show our more feasible effective than compared schemes.