北京邮电大学学报

  • EI核心期刊

北京邮电大学学报

• 学术论文 • 上一篇    下一篇

一种基于灰关联的序列模式挖掘算法

    

  1. 1.北方交通大学 交通运输学院自动化系统研究所 , 北京 100044; 2.中央民族大学, 北京 100081
  • 基金资助:
     

A Data Mining Algorithm for Sequence Pattern Based on Grey Association

    

  1. 1. Automation System Institute of Transportation Department, Northern Jiaotong University, Beijing 100044, China; 2. Central University for Nationalities, Beijing 100081, China
  • Supported by:
     

摘要: 序列模式挖掘算法多是利用了关联规则挖掘中的 Apriori特性 .利用灰关联方法对原始序列 进行净化处理 ,从而减少挖掘算法中的噪声数据 . 其理论依据在于 ,如果一个序列是频繁的 ,那么该序列的时间间隔也必然是频繁的. 利用了灰关联分析方法找出两个项之间的频繁时间间隔 ,再利用该间隔扫描事务序列数据库 ,从而最终找出频繁序列 . 

关键词: 数据挖掘, 灰关联 , 序列模式

Abstract: Most of the Sequence Pattern Mining ( SPM ) algorithm are using the Apriori characteristic of Association Rule Mining ( ARM). The paper here mostly emphasises purifying the original sequence by making use of Grey Association( GA) method to reduce the noise data during the process of mining algorithm. The academic evidence here is that if a sequence was frequent then the time intervals between every two items included in the sequence were also frequent. Therefore, firstly the paper makes use of GA method to find the frequent time interval between two items in the sequence, then according to the frequent time interval scans the affair sequence database and finally finds out the frequent sequence.

Key words: data mining, grey association, sequence pattern

中图分类号: