Self—Organizing Genetic Algorithm Based Method for Constructing Bayesian Networks from Databases

The trpical characteristic of the topology of Baycsian nctworks(BNs)is the interdependence among difffcrent nodes(variables),which makes it impossible to optimize one variablc indcpcndcntly of others,and the learning of BNs structures by general genetic algorithms is liable to converge to local extrcmum.To resolve effi-ciently this problem,a self-organizing gcnctic algorithm(SGA)based method for constructing BNs from databas-es in presented.This method makes use of a self-otganizing mechanism to dcvclop a genetic algorithm that extend-ed the crossover operalor from one to two,providing mutual competition between them,evcn adiusting the num-bers of patents in recombination(crossover/recomposition)schemes.With the K2 algorithm,this metod also optimizes the genetic operators,and utilizes adcquatcly the domain knowledge.As a result,with this method it is able to find a glohal optimum of the topology of BNs,avoiding ptcmaturc convergence to local extremum.The experimental results ptoved to be and the convergence of the SGA was discussed.

Self-Organizing Genetic Algorithm Based Method for Constructing Bayesian Networks from Databases

The typical characteristic of the topology of Bayesian networks (BNs) is the interdependence among different nodes (variables), which makes it impossible to optimize one variable independently of others, and the learning of BNs structures by general genetic algorithms is liable to converge to local extremum. To resolve efficiently this problem, a self-organizing genetic algorithm (SGA) based method for constructing BNs from databases is presented. This method makes use of a self-organizing mechanism to develop a genetic algorithm that extended the crossover operator from one to two, providing mutual competition between them, even adjusting the numbers of parents in recombination (crossover/recomposition) schemes. With the K2 algorithm, this method also optimizes the genetic operators, and utilizes adequately the domain knowledge. As a result, with this method it is able to find a global optimum of the topology of BNs, avoiding premature convergence to local extremum. The experimental results proved to be and the convergence of the SGA was discussed.