多粒度粗糙集数据分析研究的回顾与展望

Reviews and prospects on the study of multi-granularity rough set data analysis

粒计算是知识表示和数据挖掘的一个基本问题。粗糙集理论是知识获取的一个典型粒计算模型。在传统粗糙集数据分析中,信息系统中每一个对象在每一个属性上只能取唯一的值。在实际生活数据中,根据不同的粒度或者尺度,对象在同一属性上可以取不同粒度的值。该文介绍目前流行的三类基于粗糙集的多粒度数据处理模型,即多粒化粗糙集模型、多粒度邻域粗糙集模型、多尺度信息系统的粗糙集数据分析模型,回顾这三类多粒度粗糙集数据分析模型的研究进展及主要研究内容,并提出若干研究问题。

Granular computing （GrC） is a basic issue in knowledge representation and data mining. Rough set theory is a typical GrC model for knowledge acquisition.In traditional rough set data analysis, each object under each attribute in an information system can only take on one value. However, in many real-life data, objects are usually measured at different scales/granularities under the same attribute.In this paper, three types of multi-granular rough-set-data-analysis models,which are called multi-granulation rough set model, multi-granular neighborhood rough set model, and multi-scale information system based rough set data analysis model, with their research progress are reviewed. And some problems are proposed for further study.