模糊粗糙数据模型:一种数据分析的新方法

Fuzzy Rough Data Model: A New Technique for Analyzing Data

提出了一种数据分析的新方法———模糊粗糙数据模型(Fuzzy Rough Data Model,FRDM).该方法采用动态自适应模糊聚类技术,将Kowalczyk方法中的粗糙数据模型(Rough Data Model,RDM)对输入数据空间的网格状“硬划分”转化为模糊划分,辨识输入数据空间中的模糊模式类,并通过定义各模糊模式类与决策类别之间的类型映射关系ftype:Ci→y,以及输入数据对各模式类分类规则的匹配度(Degree of Fulfillment,DoF(x))概念,建立起相应的FRDM模型.不同数据集的实验测试结果表明,与Kowalczyk的RDM方法相比,文中方法具有更好的数据概括能力、更强的噪声数据处理能力和更高的搜索效率.

A new technique for analyzing data, fuzzy rough data model, FRDM, is proposed. By means of the dynamic adaptive fuzzy clustering techniques, the approach turn the grid hard partition of input data space in Kowalczyk＇s rough data model （KRDM） to the fuzzy partition, and identify the fuzzy pattern clusters of input data space. Then, the FRDM is built through utilizing the definition of type mapping relation ftype. Ci→y from each fuzzy pattern clusters to the decision categories as well as the concept DoF（x）, which is the degree of fulfillment of an input data relative to the classification rules for the pattern clusters. Finally, different experimental databases are calculated and the results demonstrate that above approach has better generalization ability, more powerful ability to handle data contaminated by noise and higher searching efficiency compared with the Kowalczyk＇s RDM.