| 基于FP树的多最小支持度广义关联规则挖掘算法 |
| |
| 引用本文： | 佘俊胜,黄战,李亚丹.基于FP树的多最小支持度广义关联规则挖掘算法[J].小型微型计算机系统,2007,28(12):2212-2215. |
| |
| 作者姓名： | 佘俊胜 黄战 李亚丹 |
| |
| 作者单位： | 暨南大学,计算机科学系,广东,广州,510632 |
| |
| 摘 要： | 采用MIS-tree结构保存频繁模式的信息提出了基于频繁模式增长挖掘原型的CFP-tax算法,该算法可避免候选集的生成和高代价的数据库扫描并能高效地找出数据库中所有频繁项集.基于虚拟数据集对算法的性能进行了评估,结果表明CFP-tax算法比经典的MMS-Cumulate算法性能有显著的提高.
|
| 关 键 词： | 数据挖掘 广义关联规则 多最小支持度 频繁模式树 |
| 文章编号： | 1000-1220（2007）12-2212-04 |
| 收稿时间： | 2006-08-18 |
| 修稿时间： | 2006年8月18日 |
Generalized Association Rule Mining Algorithm with Multiple Minimum Supports Based on Frequent Pattern Tree |
| |
| SHE Jun-sheng,HUANG Zhan,LI Ya-dan.Generalized Association Rule Mining Algorithm with Multiple Minimum Supports Based on Frequent Pattern Tree[J].Mini-micro Systems,2007,28(12):2212-2215. |
| |
| Authors: | SHE Jun-sheng HUANG Zhan LI Ya-dan |
| |
| Abstract: | A generalized association rule with multiple minimum supports mining algorithm based on pattern growth mining paradigm called CFP-tax, is proposed by employing the MIS-tree to store the information about frequent patterns, which avoids candidate generation and costly database scans and efficiently finds all the frequent item sets in the database. The performance of the algorithm is evaluated based on synthetic datasets, and the results show that our algorithm significantly outperforms the classic MMS_Cumulate algorithm. |
| |
| Keywords: | data mining generalized association rule multiple minimum supports frequent pattern tree |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
