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

中文关键词  知识化制造  知识网  模糊关联聚类  相似度  矩阵分解

英文关键词  knowledgeable manufacturing  knowledge network  fuzzy relational clustering  similarity  matrix decomposition

中文摘要

      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.

英文摘要

      The knowledge mesh’s selection method based on fuzzy relation clustering is proposed for the selecting problem 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.