An Analysis on Efficient Strategies and Applications of Spatial Data Mining |

ABSTRACT:

Spatial data mining isthe process of discovering interesting and previously unknown, but potentiallyuseful patterns from large spatial datasets. Extracting interesting and usefulpatterns from spatial datasets is more difficult than extracting thecorresponding patterns from traditional numeric and categorical data due to thecomplexity of spatial data types, spatial relationships, and spatialautocorrelation. The requirements of mining spatial databases are differentfrom those of mining classical relational databases. The spatial data miningtechniques are often derived from spatial statistics, spatial analysis, machinelearning, and databases, and are customized to analyze massive data sets. Inthis report some of the spatial data mining techniques have discussed alongwith some applications in real world. Clustering is one of themajor tasks in data mining. In the last few years, Clustering of spatial datahas received a lot of research attention. Spatial databases are components ofmany advanced information systems like geographic information systems VLSIdesign systems. In this paper, we introduce several efficient algorithms forclustering spatial data. Spatial clustering, whichgroups similar spatial objects into classes, is an important component ofspatial data mining. Spatial clustering can be used in the identification ofareas of similar land usage in an earth observation database or in merging regions with similar weather patterns,etc. As a data mining function, spatial clustering can be used as a stand-alonetool to gain insight into the distribution of data, to observe thecharacteristics of each cluster, and to focus on a particular set of clustersfor further analysis. It may also serve as a preprocessing step for otheralgorithms, such as classification and characterization, which will operate onthe detected clusters. Spatial data mining isthe discovery of interesting relationships and characteristics that may existimplicitly in spatial databases. In this paper, we explore whether clusteringmethods have a role to play in spatial data mining.