| 基于投影数据集的序列模式增量挖掘算法 |
| |
| 引用本文： | 蒋栋栋,孙志挥.基于投影数据集的序列模式增量挖掘算法[J].计算机工程与应用,2006,42(26):156-158,166. |
| |
| 作者姓名： | 蒋栋栋 孙志挥 |
| |
| 作者单位： | 东南大学计算机科学与工程系,南京,210096 |
| |
| 摘 要： | 提出一种基于投影数据集的序列增量更新算法Inc_SPM,该算法以PrefixSpan算法为基础。首先利用已有的知识得出频繁1序列,然后生成投影数据集以迭代产生频繁k序列;同时为了控制投影数据集的规模,利用等价投影数据集来改进投影终止条件。
|
| 关 键 词： | 数据挖掘 序列模式 增量更新 投影数据集 |
| 文章编号： | 1002-8331-（2006）26-0156-03 |
| 收稿时间： | 2006-04 |
| 修稿时间： | 2006-04 |
Incremental Update Algorithm of Sequential Patterns Mining Based on Projected Datasets |
| |
| JIANG Dong-dong,SUN Zhi-hui.Incremental Update Algorithm of Sequential Patterns Mining Based on Projected Datasets[J].Computer Engineering and Applications,2006,42(26):156-158,166. |
| |
| Authors: | JIANG Dong-dong SUN Zhi-hui |
| |
| Affiliation: | Department of Computer Science and Engineering,Southeast University,Nanjing 210096 |
| |
| Abstract: | Sequential pattern mining is an important data mining problem with broad application.The paper discusses the incremental update of frequent sequences,and a new algorithm Inc_SPM(Incremental update of Sequential Pattern Mining) based on the algorithm of PrefixSpan is provided.The algorithm utilizes previous mining results to get frequent 1-sequences,then produces projected datasets to get frequent k-sequences iteratively.To reduce the size of projected datasets,only frequent items of frequent 1-sequences are kept in projected datasets,and equivalence of projected datasets are utilized for the early termination of projection. |
| |
| Keywords: | data mining sequential pattern incremental update projected datasets |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
