Study On Social Cognitive Optimization's Improvements And Its Application

By the means of simulating evolution of the nature, especially evolution of living things, evolutionary computation (EC) is a wonderful computation model to solve complex problems. Evolutionary computation not only has feature of auto-study, auto-organized, auto-adapt, but also, has the advantage of simple, universal, and suit parallel press. Evolutionary computation is widely applied in practical applications need massive search and optimization.Society Cognitive Optimization (SCO) is such kind of evolutionary computation of simulation to human society. SCO is an intelligent optimization algorithm basing on Society Cognitive Theory (SCT). SCT thought personality was composed with heredity, environment and cognition. It takes environment and cognition the important factors, and emphasizes the significance of social learning. SCO simulates the social learning capability in SCT Through tournament selection and neighborhood searching. Agent representing person and library representing knowledge, through the interaction between agent and library, SCO simulates the human's learning process and achieves the optimization.Self-Organizing Migrating Algorithm (SOMA) is also a kind of swarm intelligence. It simulates behavior of group of animal when they are looking for food. All members in swarm search towards the best one by small jumps. In the evolvement, SOMA is only the process that individual searches towards the best one in search space, and it's not like the classic EA (two parents create one new individual - offspring).This paper introduces SOMA into SCO. Through joining SOMA's migration in the process of SCO and adding two parameters in the improved algorithm, SCO is improved. The experiment result showed the improved SCO may obtain the quick convergence rate in the optimization early and only small effect on the last result.And then, the improved SCO is applied to solve the SAT problem. The satisfiability problem of conjunction normal form (abbreviate SAT problem) is the core problem of contemporary theory computer science and it's a typical NP-complete problem. Massive important applications of modern science, technology, military and the economical management all sum up for solving the NP complete question, so the fast solution not only has important academic significance, but also has important...