文档介绍：Data Mining for Selective Visualization of Large Spatial Datasets
Ý
Shashi Shekhar ,£ Chang-Tien Lu , Pusheng Zhang, Rulin Liu
Computer Science & Engineering Department
University of Minnesota
Email: [shekhar, ctlu, pusheng, rliu]***@
Abstract useful but implicit knowledge in spatial databases. With
the huge amount of spatial data obtained from satellite im-
Data mining is the process of extracting implicit, valu- ages, medical images, and geographical information sys-
able, and interesting information from large sets of data. tems (GIS), it is a non-trivial task for humans to explore spa-
Visualization is the process of visually exploring data for tial data in detail. Spatial datasets and patterns are abundant
pattern and trend analysis, and it is mon method in many application domains related to NASA, the Environ-
of browsing spatial datasets to look for patterns. How- mental Protection Agency, the National Institute of Stan-
ever, the growing volume of spatial datasets make it diffi- dards and Technology, and the Department of Transporta-
cult for humans to browse such datasets in their entirety, tion. A key goal of spatial data mining is to partially auto-
and data mining algorithms are needed to filter out large mate knowledge discovery, ., search for “nuggets” of in-
uninteresting parts of spatial datasets. We construct a formation embedded