论文部分内容阅读
属性约简是数据挖掘的一个重要研究内容.为了解决具有多种属性类型的决策表约简问题,在粗集和二元关系聚合理论的基础上,利用属性重要性作为评价标准,提出了一种两阶段遗传约简算法.算法的第一阶段是为了找出尽可能多的约简,第二阶段力求寻找最小约简.根据算法每个阶段的目标设计了编码方案、种群规模、适应度函数、终止条件、选择、变异和修正操作.实验表明,与标准遗传算法相比,两阶段算法在计算最小约简时更为准确和稳定.
Attribute reduction is an important research content of data mining.In order to solve the problem of decision table reduction with many attribute types, on the basis of rough set theory and binary relation aggregation theory, using attribute importance as evaluation criteria, The first phase of the algorithm is to find out as many reductions as possible and the second phase seeks to find the minimum reduction.According to the target of each stage of the algorithm, the coding scheme, population size, fitness Function, termination condition, selection, mutation and rectification operation.Experiments show that compared with standard genetic algorithm, the two-stage algorithm is more accurate and stable when calculating the minimum reduction.