数据流中的频繁标记闭子树的批量挖掘

doi:10.13190/jbupt.201005.121.367

北京邮电大学学报 ›› 2010, Vol. 33 ›› Issue (5): 121-125.doi: 10.13190/jbupt.201005.121.367

数据流中的频繁标记闭子树的批量挖掘

冯博，徐雅静，赵娜，徐惠民

北京邮电大学信息与通信工程学院

收稿日期:2009-12-09 修回日期:2010-01-28 出版日期:2010-08-28 发布日期:2010-06-15
通讯作者: 冯博 E-mail:emoskyny@gmail.com
基金资助:
高等学校学科创新引智计划(B08004);国家级.国家自然科学基金项目;国家级.国家高技术研究发展计划项目

A Batch Mining Algorithm for Frequent Closed Labeled Trees in Data Streams

Received:2009-12-09 Revised:2010-01-28 Online:2010-08-28 Published:2010-06-15

摘要/Abstract

摘要：

在频繁模式挖掘（FPM）的研究中，为了在海量数据流中有效地挖掘子树结构的频繁模式，根据数据流和子树模式的特点，提出了一种基于数据流的频繁标记闭子树挖掘（SFCLTreeMiner）算法. 该算法首次对动态数据流中频繁标记闭子树的挖掘进行研究,给出了在数据流中标记闭子树集合添加、删除的批量挖掘方法,并结合时间衰减模型,有效保证了结果的时效性. 实验结果表明,该算法在挖掘性能，如挖掘时间和内存占用等方面，比类似算法有较大提高.

关键词: 频繁子树挖掘, 数据流, 时间衰减, 标记闭子树, 批量挖掘

Abstract:

Compared with the classic frequent pattern mining (FPM) algorithms, the dynamic FPM algorithms on fast and massive data streams have become top research nowadays. A new batch mining algorithm in data streams called stream frequent closed labeled tree miner (SFCLTreeMiner) is proposed. SFCLTreeMiner uses a kind of addingremoving method between closed tree sets. Also it provides a time decay module for reasonable data updating. Experiment shows that SFCLTreeMiner is efficient in data streams mining by reducing consuming dramatically.

Key words: frequent subtree mining, data streams, time decay, labeled closed subtree, batch mining

中图分类号:

TP311

冯博，徐雅静，赵娜，徐惠民. 数据流中的频繁标记闭子树的批量挖掘[J]. 北京邮电大学学报, 2010, 33(5): 121-125.

参考文献

[1] 徐前方, 肖波, 郭军. 一种基于相关度统计的告警关联规则挖掘算法[J]. 北京邮电大学学报, 2007, 29(1): 63-66. Xu Qianfang, Xiao Bo, Guo Jun. A mining algorithm with alarm association rules based on statistical correlation[J]. Journal of Beijing University of Posts and Telecommunications, 2007, 29(1): 63-66.
[2] Zaki M J. Efficiently mining frequent trees in a forest: algorithms and application[J]. IEEE Trans on Knowledge and Data Engineering, 2005, 17(8): 1021-1035.
[3] Chi Yun, Yang Yirong, Xia Yi, et al. CMTreeminer: mining both closed and maximal frequent subtrees// PAKDD04. Sydney: , 2004: 63-73.
[4] Termier A, Rousset M C, Sebag M, et al. DryadeParent: an efficient and robust closed attribute tree mining algorithm[J]. IEEE Trans on Knowledge and Data Engineering, 2008, 20(3): 300-320.
[5] Chi Yun, Wang Haixun, Yu P S, et al. Moment: maintaining closed frequent itemsets over a stream sliding window//ICDM04. Brighton: , 2004: 59-66.
[6] 李国徽, 陈辉. 挖掘数据流任意滑动时间窗口内频繁模式[J]. 软件学报, 2008, 19(10): 2585-2596. Li Guohui, Chen Hui. Mining the frequent patterns in an arbitrary sliding window over online data streams[J]. Journal of Software, 2008, 19(10): 2585-2596.
[7] Leung C K S, Khan Q I. DSTree: a tree structure for the mining of frequent sets from data streams// ICDM06. Hong Kong: , 2006: 928-932.
[8] Hwang J H, Gu M S. Finding frequent structures in XML stream data// CSA 2009. Jeju: , 2009: 3-8.
[9] Bifet A, Gavalda R. Mining adaptively frequent closed unlabeled rooted trees in data streams// KDD 2008. Las Vegas: , 2008: 34-42.
[10] Chalmers R, Almeroth K. Modeling the branching characteristics and efficiency gains of global multicast trees// IEEE INFOCOM 2001. Anchorage:, 2001: 449-458.
[11] Han Jiawei, Kamber M. Data mining concepts and techniques[M]. 2nd ed. San Francisco: Morgan Kaufmann, 2006.

数据流中的频繁标记闭子树的批量挖掘

A Batch Mining Algorithm for Frequent Closed Labeled Trees in Data Streams

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

编辑推荐

Metrics

本文评价

[1]	苏静, 裘晓峰, 李书芳, 刘道伟, 张春红. 一种基于原型学习的自适应概念漂移分类方法[J]. 北京邮电大学学报, 2017, 40(3): 43-50.
[2]	赵金东, 于彦伟, 刘惊雷. 面向实时海量数据流的数据聚类[J]. 北京邮电大学学报, 2016, 39(3): 114-119.
[3]	唐伦, 刘益富, 刘青海, 陈前斌. 一种改进IEEE 802.11 DCF的建模与分析[J]. 北京邮电大学学报, 2014, 37(5): 96-99.
[4]	张健沛,杨显飞,杨静. 面向高速数据流的偏倚抽样集合分类器[J]. 北京邮电大学学报, 2010, 33(4): 44-48.
[5]	张雷1,肖丁1,艾波2. 面向电信运营流程的数据稽核方法[J]. 北京邮电大学学报, 2006, 29(s1): 126-129.
[6]	景晓军，李剑锋，张苏慧，杨义先. 图像智能化的目标检测技术(Ⅱ)——数据流程与背景感知[J]. 北京邮电大学学报, 2003, 26(S1): 17-27.
[7]	刘云龙,陈俊亮. 一种新的软件容错方法及应用^*[J]. 北京邮电大学学报, 1998, 21(1): 23-28.