区间值数据库上的正态关联规则

Normal Association Rules on Interval Valued Database

通过等距处理在数据库的区间值上取样 ,应用模糊 c-方法算法确定正态模糊数的两个参数 ,借助正态模糊数模型来软化数量属性论域的划分边界。给出正态关联规则的挖掘方法与预测方法。由于文中的方法能将数量属性的正态关联规则的问题转化为布尔属性的关联规则的问题 。

In this paper, samples are selected through equally spaced treatments on interval values of database and used to determine two parameters of normal fuzzy numbers by fuzzy c-means algorithm. Then the normal fuzzy number model is adopted to soften the domain partition boundary of the quantitative attributes. The mining and prediction methods of the normal association rules are also provided. In these methods, the issue of quantitative normal association rules can be translated into the issue of binary association rules, so it can be easily understood and grasped.