摘要
针对数据流的特点,提出利用函数迭代运算的方法来提取数据流中的频繁项集的挖掘方法.整个挖掘过程只需扫描数据流一次,不产生频繁候选集.算法的时间复杂度是比较低的.实验仿真结果也验证了该挖掘方法是有效的和可行的.
In this paper,basing on the characteristics of data stream,we propose a mining approach which uses a method of function iterative to excavate the frequent itemsets hidden in the data stream.This algorithm only needs one scan over the data stream,and does＇t bring about any frequent candidate itemsets.The mining algorithm has the performance of a low time complexity.It also shows from our simulation experiment that this mining approach is effective and feasible for mining the data stream.
出处
《广西民族大学学报（自然科学版）》
CAS
2012年第1期45-49,共5页
Journal of Guangxi Minzu University ：Natural Science Edition
基金
广西自然科学基金(0832084)
广西混杂计算与集成电路设计分析重点实验室~~
关键词
数据挖掘
数据流
函数迭代
频繁项集
data mining
data stream
function iterative operation
frequent itemset
