| 闭合序列模式挖掘算法 |
| |
| 引用本文： | 沙金,邓成玉,张翠肖,刘伟峰.闭合序列模式挖掘算法[J].计算机工程与设计,2006,27(3):514-518. |
| |
| 作者姓名： | 沙金 邓成玉 张翠肖 刘伟峰 |
| |
| 作者单位： | 1. 石家庄铁道学院,计算机系,河北,石家庄,050043
2. 燕山大学,信息工程学院,河北,秦皇岛,066004 |
| |
| 摘 要： | 提出了一种新的挖掘闭合序列模式的PosD算法，该算法利用位置数据保存数据项的顺序信息，并基于位置数据列表保存数据项的顺序关系提出了两种修剪方法：逆向超模式和相同位置数据。为了确保栅格存储的正确性和简洁性，另外还针对一些特殊情况做处理。试验结果表明，在中大型数据库和小支持度的情况下谊算法比CloSpan算法更有效。
|
| 关 键 词： | 数据挖掘 序列模式 闭合序列模式 逆向超模式 |
| 文章编号： | 1000-7024（2006）03-0514-05 |
| 收稿时间： | 2004-12-29 |
| 修稿时间： | 2004-12-29 |
Closed sequential pattern mining algorithm |
| |
| SHA Jin,DENG Cheng-yu,ZHANG Cui-xiao,LIU Wei-feng.Closed sequential pattern mining algorithm[J].Computer Engineering and Design,2006,27(3):514-518. |
| |
| Authors: | SHA Jin DENG Cheng-yu ZHANG Cui-xiao LIU Wei-feng |
| |
| Affiliation: | 1. Department of Computer Science and Technology, Shijiazhuang Railway Institute, Shijiazhuang 050043, China; 2. College of Communication Engineering, Yanshan University, Qinhuangdao 066004, China |
| |
| Abstract: | A new closed sequential pattern mining algorithm was proposed. PosD algorithm used a list of positional data to reserve information of item ordering. By using these positional data, two main pruning techniques, backward super-pattern condition and same positional data condition were developed. To ensure correct and compact resulted lattice, some special conditions were manipulated. From experimental results, this algorithm outperforms CloSpan in cases of moderately large datasets and low support threshold. |
| |
| Keywords: | data mining sequential pattern closed sequential pattern backward super-pattern |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
