摘要
首先,通过分析程度多粒度粗糙集和加权多粒度粗糙集的特点,针对不完备信息系统提出基于限制容差关系的加权程度多粒度粗糙集模型.然后提出一种粒度约简方法,建立粒度矩阵获得核粒度,并以粒度重要性作为启发式信息选择粒度,获得最终的约简集.最后,通过实例分析验证所提出的模型和粒度约简算法的有效性.
First of all, by analyzing the characteristics of grade multi-granulation rough set and weighted multi-granulation rough set, a weighted grade multi-granulation rough set based on limited tolerance relation about incomplete information system is presented. Then a method of granulation reduction is proposed. By establishing the granularity matrix to obtain nuclear granularity and regarding the importance of granularity as heuristic information to select granularity, the final reduction set is obtained. Finally, the results of an example also show the effectiveness of the model and method of granulation reduction.
出处
《小型微型计算机系统》
CSCD
北大核心
2017年第10期2286-2290,共5页
Journal of Chinese Computer Systems
基金
国家青年科学基金项目(61300059)资助
安徽省高校自然科学研究重大项目(KJ2014ZD05)资助
安徽高校自然科学研究项目(KJ2016A098)资助
关键词
不完备信息系统
限制容差关系
多粒度粗糙集
粒度约简
incomplete information system
limited toleration relations
multi-granulation rough set
granular reduction
