基于粗糙集的海量数据挖掘

文章结合粗糙集在处理海量数据的约简方面的优势和层次分析法的决策优势，建立了以决策属性为上层因素，紊件属性为下层因素的合理评价模型，并对提取的规则再次进行了约简。文中利用改进的粗糙集属性约简算法来降低海量数据的冗余度，提取约简后的规则，借助粗糙集的属性重要度理论．弥补了层次分析法中评价因子的主观因素。此算法模型省略对核的提取过程，对提取的规则进行了定量的分析，实现了海量数据在属性与规则上的约简。实例证明了算法的有效性。

On magnanious Data Redundanly Based on Rough Set

This paper establishes a reasonable evaluation model with the decision-making attribute as the upper factor while the condition attribute as the lower factor,considering both the advantage of reduction in processing massive data and the AHP advantage of the decision-making in Rough Set theory.And it also does a further reduction to the rules extracted.The paper utilizes the attribute reduction algorithm improved to reduce the magnanimous data redundancy,extracts the rules after reduction.With the aid of the...