Abstract:
In order to improve search efficiency of data stream frequent closed itemsets, this paper proposes an improved NewMoment algorithm to mine frequent closed itemsets over data streams. By adding level node in LevelCET data structure and using level checking strategy and optimum frequent closed items, it can quickly tap all the frequent closed itemsets over data streams. Expertimental results show the improved algorithm is better than NewMoment.
Key words:
data stream,
frequent closed itemset,
sliding window,
NewMoment algorithm,
LevelCET data structure
摘要： 为提高数据流频繁闭项集的查找效率，提出一种改进的NewMoment频繁闭项集挖掘算法，通过在LevelCET数据结构中加入层次结点，并利用层次检测策略与最佳频繁闭项集检测策略快速挖掘数据流滑动窗口中所有的频繁闭项集。实验结果证明，与NewMoment算法相比，改进的算法性能更优。
关键词:
数据流,
频繁闭项集,
滑动窗口,
NewMoment算法,
LevelCET数据结构
CLC Number:
LIU Ji, YANG Lu-Meng, MAO Yi-Min, LIU Li-Xin, XIE Dong. Improved Mining Algorithm for Frequent Closed Itemsets of Data Stream[J]. Computer Engineering, 2011, 37(9): 75-77.
刘洁, 杨路明, 毛伊敏, 刘立新, 谢东. 改进的数据流频繁闭项集挖掘算法[J]. 计算机工程, 2011, 37(9): 75-77.