地理科学进展 ›› 2007, Vol. 26 ›› Issue (6): 123-132.doi: 10.11820/dlkxjz.2007.06.013

• 农业与生态 • 上一篇    下一篇

空间关联规则挖掘研究进展

张雪伍1,2, 苏奋振1, 石忆邵2, 张丹丹1   

  1. 1. 中国科学院地理科学与资源研究所资源与环境信息系统国家重点实验室, 北京100101|
    2. 同济大学测量与国土信息工程系, 上海200092
  • 收稿日期:2007-08-01 修回日期:2007-11-01 出版日期:2007-11-25 发布日期:2007-11-25
  • 作者简介:张雪伍(1981-), 男, 安徽泗县人, 同济大学博士研究生, 主要研究方向为时空数据挖掘及应用. E-mail: zhangxuewu4116@163.com
  • 基金资助:

    国家重点基础研究发展计划资助项目( 2006CB701305) ; 国家自然基金项目( 40571129) .

Resear ch on Progr ess of Spatial Association Rule Mining

ZHANG Xuewu1,2, SU FenZhen1, SHI Yishao2, ZHANG Dandan1   

  1. 1. Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China|
    2. Department of Surveying and Geo- informatics, Tongji University, Shanghai 200092, China
  • Received:2007-08-01 Revised:2007-11-01 Online:2007-11-25 Published:2007-11-25

摘要:

随着空间数据获取技术的进步, 空间数据量日益增大, 已超出人们的分析能力。传统的空 间数据分析方法只能进行简单的数据分析, 无法满足人们获取知识的需要。空间关联规则是空间 数据挖掘一个基本的任务, 是从具有海量、多维、多尺度、不确定性边界等特性的空间数据中进行 知识发现的重要方法。本文从基本概念、分类、挖掘过程、挖掘方法、目前研究成果等方面对其进 行综述, 重点阐述了空间关联规则挖掘效率的改进策略、基于不确定空间信息的挖掘方法、挖掘 过程及结果的可视化、弱空间关联规则的挖掘方法等。通过对现有空间关联规则研究成果和存在 问题的深入剖析, 指出了其未来主要的发展方向。

关键词: 地理信息系统(GIS), 空间关联规则, 空间数据, 数据挖掘

Abstract:

With the progress of spatial data technologies, the volumes of the spatial data enhance gradually, far exceeding people’s ability to analyze it. Traditional spatial data analysis methods can only carry out simple data analysis, having no way to satisfy people’s need of gaining knowledge. Spatial association rule mining approach, which is used to acquire underlying spatial knowledge from spatial database managing complex, multiple- dimension, large, and flexibility border space spatial data, is a fundamental mission of the spatial data mining. The authors make an annotated review of basic concepts, classification, mining process, current research achievements and so on, especially paying attention to approaches improving mining efficient, mining method based on the uncertain space information and attribute information, visualization of mining results and processes, and negative spatial association rule mining. After deeply analyzing research achievements and existing problems, the authors bring forward the future main development directions of spatial association rule mining.

Key words: association rule mining, data mining, geographic information system, spatial data