期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2008
卷号：XXXVII Part B2
页码：225-228
出版社：Copernicus Publications
摘要：For the sake of environmental change monitoring, a huge amount of geospatial and temporal data have been acquired through various networks of monitoring stations. For instance, daily precipitation and air temperature are observed at meteorological stations, and MODIS images are regularly received at satellite ground stations. However, so far these massive raw data from the stations are not fully utilized, or say, geographical spatio-temporal structural information in raw data aren't exposed sufficiently. Upon the requirements of human decision-making, explosive raw data is embarrassed in contrast to starved information or knowledge. This paper makes a short introduction to our research project, i.e., mining association rules in vegetation and climate changing data of northeastern China. We describe a framework of mining association rules in regional vegetation and climate-changing data, in which the methods of Kriging interpolation, wavelets multi-resolution analysis, fuzzy c-means clustering, and Apriori-based logical rules extraction are used respectively. Then, we give the definitions of geographical spatio-temporal transactions and multi-level fuzzy association rules, which play the decisive roles in association rules mining. It is noted that the largest difference between spatial data mining and general data mining is computation of geographical spatio-temporal correlations or variability. The whole procedure of geographical spatio-temporal data mining can be thought of as multi-stage computation of geographical spatio-temporal correlations. At the end of this paper, towards an advanced regional vegetation-climate changing data mining system, several underlying techniques of mining spatio-temporal association rules are initiatively pointed out for next work
关键词：Vegetation; Climate; Multi-level; Fuzzy; Association Rules; Spatio-temporal; Data Mining
