多属性修正遗传算法适应度在专家系统中的应用  

Application of Fitness of Multiattribute Fixed Genetic Algorithm in the Expert System

作　　者：张卫[1] 彭月英[1] 魏勇[1] 冯振[1] 

机构地区：[1]广西师范学院计算机与信息工程学院,广西530023

出　　处：《现代计算机》2011年第11期25-29,共5页Modern Computer

摘　　要：遗传算法算法原理简单明了,在计算极值方面有其竞争优势,可以方便地逃离局部最优解达到全局最优解。算法运行的核心是适应度函数,起引导算法优化指向的作用。提出一种多属性遗传算法适应度描述算法,从而使得现实中的推理问题可以利用遗传算法的求优能力来解决,试验证明它具有很强的可行性和实用性。The genetic algorithm algorithm principle is simply, and has the competitive advantage in the computation extreme value aspect. May facilitate flees the partial optimal solution to achieve the globally optimal solution. The algorithm movement＇s core is the sufficiency function, and has the function that guides the direction of algoritym optimization. Proposes a kind of muhiattribute genetic sufficiency description algorithm, thus causes the inference question in the reality to be solved by using the superior ability of genetic algorithm, and the experiment proves that it has the very strong feasibility and usability.

关 键 词：遗传算法 多属性 推理 极值 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]