基于模糊关联聚类的知识网选择方法  

Selection of knowledge mesh based on fuzzy relational clustering

作　　者：杨人子[1,2] 严洪森[1,3] 

机构地区：[1]东南大学自动化学院,江苏南京210096 [2]东南大学数学系,江苏南京210096 [3]复杂工程系统测量与控制教育部重点实验室,江苏南京210096

出　　处：《控制理论与应用》2013年第1期8-16,共9页Control Theory & Applications

基　　金：国家自然科学基金重点资助项目(60934008);江苏省博士后基金资助项目(1202102C)

摘　　要：针对知识化制造系统中相似知识网日益增多和用户需求表达不清晰等导致的知识网选择问题,提出一种基于模糊关联聚类的知识网选择方法.综合知识网功能、完善程度和结构等方面构造的相似度具有反映知识网运算规律的特征.将两两知识网的相似度作为聚类数据,降低了高维特征空间的维数.模糊关联矩阵的分解,获得了知识网--类关系.目标知识网与类中类隶属度高的知识网的比较缩小了用户选择范围.最后的实例表明该方法是有效可行的.The knowledge mesh＇s selection method based on fuzzy relation clustering is proposed for the selecting prob- lem caused by the increasing similar knowledge mesh and unclear requirements of user in knowledgeable manufacturing system. The similarity degree which synthesizes the function, perfection degree and structure of knowledge mesh, has the characteristics of reflecting operation laws. The similarity values as cluster data reduce the dimension of high-dimensional feature space. The decomposition of fuzzy relational matrix obtains the groups of knowledge meshes. The comparisons between target knowledge mesh and knowledge mesh with high class membership degree in each class narrow the scope of user selection. The last example shows that the method is effective and feasible.

关 键 词：知识化制造 知识网 模糊关联聚类 相似度 矩阵分解 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]