摘要
提出了双重区间值聚类的数据挖掘模型 :首先将每个属性的取值按照领域知识划分为若干类 ,接着统计每个类在各条“交易”中出现的频率 (支持度 ) ,最后再按照关联规则挖掘方法进行处理 .这种区间值数据挖掘方法与传统的数据挖掘方法相比较 。
A new model for data mining named double-interval clustering is brought forward.Firstly,divide the values of each attribute into some classes;secondly,sum the frequencies (supports) of every transaction item appearing in all different classes;finally,according to the association rule mining methods handle the data.This kind of interval data mining way has more practical value compared with the traditional one.So it has a larger free development space.
出处
《广西师范大学学报(自然科学版)》
CAS
2004年第3期15-18,共4页
Journal of Guangxi Normal University:Natural Science Edition
基金
澳大利亚 ARC基金资助项目 ( DP0 3 43 1 0 9)