基于商空间的模糊聚类方法研究

Research of Fuzzy clustering based on quotient space

商空间粒度理论给我们提供了一种研究不同粒度世界的方法,通过粗细粒度的转换,能够满足不同层次的需求。本文提出一种基于商空间的模糊聚类分析模型,结合了商空间粒度理论、分层递阶结构和模糊综合评价模型的思想。最后给出应用实例对客户进行细分,实验证明该模型在解决多维数据分类问题时是有效合理的。

Quotient space theory offers us one kind of method to represent the world at different granularities,it can meet different requirements through the transform of coarse granularity to fine granularity. In this paper,we proposed a model of Fuzzy cluster analysis based on quotient space. It covers the ideas of quotient space theory,hierarchical structure and fuzzy synthetic evaluation. At last we offer an application of subdivision on client. The application shows that this model is reasonable and efficient on dealing with classification of multi-dimension data.