摘要
针对传统空间数据关联规则挖掘缺乏不确定性处理及度量的局限性,将空间数据的不确定性和空间数据挖掘的不确定性有机结合,初步建立了空间数据关联规则挖掘的不确定性处理模型及度量指标,包括空间数据不确定性的Monte Carlo模拟、基于不确定性空间数据的空间自相关度量和关联规则不确定性度量等,并以我国某地区环境调查数据为例进行验证。
In order to overcome the deficiencies of traditional spatial data association rules mining that is short of uncertainty processing and measurement, the uncertainties of spatial data and spatial data mining were properly combined and uncertainty processing model and measurement indexes of spatial data association rules mining had been founded. In which, four key problems had been probed and analyzed, including uncertainty simulation of spatial data with Monte Carlo methods, measurement of spatial autccorrelation based on uncertain spatial positional data, discreteness of continuous data based on uncertain spatial clustering algorithm and uncertainty measurement of association results. Meanwhile, the experiences concerned were performed using the geo - spatial environment data gotten from one area in China.
出处
《地理与地理信息科学》
CSCD
北大核心
2006年第6期5-8,共4页
Geography and Geo-Information Science
基金
国家自然科学基金项目"空间数据挖掘的若干理论与技术问题的研究"(60275021)
中国博士后科学基金项目"多传感器卫星遥感数据产品四维同化的时空尺度及不确定性研究"(20060390326)
关键词
不确定性
空间数据挖掘
关联规则
uncertainty
spatial data mining
association rules
