摘要
运用选择、投影、广义笛卡尔积等关系代数运算 ,给出了遗传算法的搜索空间及个体、遗传算子和搜索最优解过程等关系代数形式的描述 ,建立了遗传算法的关系代数模型 ,给出了遗传算法的数学解释 .然后 ,给出建立遗传算法关系代数模型的意义 ,说明了数据挖掘和知识发现应用于遗传算法的可行性 .最后 ,用该模型描述了 2个常见用遗传算法解决的问题 ,即TSP问题和交互式遗传算法中的服装设计问题 ,结果表明该模型的可行性 .
The search spa ce, individuals, genetic operators and the process of searching optima in geneti c algorithm are formally described by relation algebra operators such as selecti on, projection, and generalized Cartesian product. The relation algebra based ge neti c algorithm model is established, which serves as the explanation of genetic alg o rithm. Then the significance is presented which provides the feasibility of appl ication of data-mining and knowledge-discovering to genetic algorithm. At last , the model is applied to the description of two problems solved by genetic algo rithm, i.e., travelling salesman problem (TSP) and fashion design problem in int eractive genetic algorithm (IGA). The result indicates its feasibility.
出处
《东南大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2004年第B11期58-62,共5页
Journal of Southeast University：Natural Science Edition
基金
国家自然科学基金资助项目 (60 3 0 40 16) .
关键词
遗传算法
关系代数
基因意义单元
编码
genetic algorithm
relation algebra
genetic sense unit
encoding
