基于差异关系的变精度粗糙集知识约简算法研究

Research on Knowledge Reduction Algorithm Based on Variable Precision Tolerance Rough Set Theory

有效的知识约简算法是粗糙集理论的重要研究内容。粗糙集是一个去掉冗余特征的有效工具。经典的粗糙集方法要求数值用离散数据表达,对于连续值则在处理前必须进行离散化处理。真实数据往往存在连续值,为了避免运用粗糙集方法所必需的离散化过程带来的信息丢失,将差异关系应用于粗糙集的知识约简。为进一步增强差异关系粗糙集对噪声数据的适应能力,提出基于差异关系的变精度粗糙集知识约简算法,并分析差异关系下变精度粗糙集模型参数的特性,给出依赖度和参数范围关系描述,将参数取值从点扩展到区间范围。在UCI数据库的数据集上进行实验,结果证明了所提方法及相关理论的有效性。

Knowledge reduction is an important research issue in rough set theory. Rough set theory is an efficient mathematical tool for further reducing redundancy. The main limitation of traditional rough set theory is the lack of effective methods for dealing with real-valued data. However, practical data sets are always continuous. This has been ad- dressed by employing discretization methods, which may result in information loss. This paper investigated one approach combining tolerance relation together with rough set theory. In order to enhance the ability to adapt to the noise data, this paper explored the knowledge reduction algorithm based on variable precision tolerance rough set theory. The characteristics of parameter were analyzed. The relationship between the classification quality and parameter interval was described, and the parameter value was extended to interval range. The experimental results demonstrate that our proposed algorithm and the related theory are effective.