论文部分内容阅读
在知识发觉中遗传算法已经广泛应用于分类,模型选择和其它优化问题.但是它的行为和表现却直接受其输入参数值(如交叉概率和变异概率)的影响,不合理的参数设置通常会导致许多问题比如早熟问题.为此有的学者提出用自适应技术在算法过程中自适应调整这些参数,但这并未对遗传算法产生整体的改善,因为参数设置是依赖于具体问题的.提出了基于染色体个体寿命特征的遗传算法,用模糊逻辑控制器自适应调整交叉概率和变异概率.这个方法加强了遗传算法的全局搜索能力,很好的解决了早熟问题.将本算法和标准遗传算法及自适应遗传算法比较,仿真结果表明本算法在克服早熟问题上的明显优势.
In knowledge discovery, genetic algorithms have been widely used in classification, model selection and other optimization problems, but their behavior and performance are directly affected by the input parameter values (such as crossover probability and mutation probability). The unreasonable parameter setting usually Leading to many problems such as premature problems.For this reason, some scholars have proposed using adaptive technology to adaptively adjust these parameters in the algorithm process, but this does not produce an overall improvement of genetic algorithm, because the parameter settings are dependent on the specific problem. Based on the genetic algorithm of individual life characteristics of chromosomes, the fuzzy logic controller is used to adaptively adjust the crossover probability and mutation probability.This method enhances the global search ability of genetic algorithm and solves the premature problem well.This algorithm and the standard genetic algorithm Compared with adaptive genetic algorithm, the simulation results show that this algorithm has obvious advantages in overcoming premature problems.