计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (5): 190-193.

• 数据库与信息处理 • 上一篇    下一篇

基于Cache的数据挖掘专用数据访问组件

黄晓霞1,程 论2,黄有方3   

  1. 1.上海海事大学 计算机系,上海 200135
    2.同济大学 控制科学与工程系,上海 200092
    3.上海海事大学 物流研究中心,上海 200135
  • 收稿日期:2007-05-29 修回日期:2007-07-25 出版日期:2008-02-11 发布日期:2008-02-11
  • 通讯作者: 黄晓霞

Cache-based data access component for data mining algorithm

HUANG Xiao-xia1,CHENG Lun2,HUANG You-fang3   

  1. 1.Department of Computer,Shanghai Maritime University,Shanghai 200135,China
    2.Department of Control Science and Engineering,Tongji University,Shanghai 200092,China
    3.Logistics Research Center,Shanghai Maritime University,Shanghai 200135,China
  • Received:2007-05-29 Revised:2007-07-25 Online:2008-02-11 Published:2008-02-11
  • Contact: HUANG Xiao-xia

摘要: 介绍了数据挖掘算法的两种传统数据访问方式及其缺点,提出了新的数据访问方式——基于Cache的数据挖掘算法的数据访问方法,该方法提供了三种模式缓存数据:单列模式、多列模式、混合模式,以适用多种数据挖掘算法的需要。设计实现了这种数据挖掘专用数据访问组件,该组件兼顾了传统访问方式的优点,实验证明在占用有限系统资源的情况下,保证了高效的数据访问效率并支持对海量数据的访问。

关键词: Cache, 数据访问, 数据挖掘, 组件

Abstract: This paper introduces the disadvantages of two traditional data access modes for data mining algorithms,and then presents a new Cache-based data access mode.The mode supports three buffer patterns,which are single column pattern,multi-column pattern,mixed-columns pattern.These patterns are suitable for most of the data mining algorithms.Design and implementation of this Cache-based data-mining-specific data access component is presented later.It takes advantages of two traditional modes,ensures the high performance and supports the great capacity of data access with limited system resource.

Key words: Cache, data access, data mining, component