Spatial data mining denotes the extraction of spatial patterns from both spatial and aspatial data, possibly stored in a spatial database. An important application area where spatial data mining techniques can be effectively used is offered by GIS. This paper presents an extension of a prototypical GIS, named INGENS (Inductive Geographic Information System), which integrates data mining tools to support both predictive (i.e., classification) and descriptive (i.e., association discovery) tasks. The entire data mining process is devised for a user who controls the parameters of the process by means of a query written in SDMOQL, a spatial mining query language, whose design is based on the standard OQL. Five primitives have been considered as guidelines for the design of SDMOQL, namely, the set of task-relevant data to be mined, the kind of knowledge to be mined, the background knowledge to be used in the discovery process, the interestingness measures and thresholds for pattern evaluation, and the expected representation for visualizing the discovered patterns. We illustrate an application of SDMOQL to the task of Apulia map interpretation in order to discover classification rules of interest to town planners.
Extending GIS with a Spatial Data Mining Query Language: A Case Study
APPICE, ANNALISA;ESPOSITO, Floriana;LANZA, Antonietta;MALERBA, Donato
2003-01-01
Abstract
Spatial data mining denotes the extraction of spatial patterns from both spatial and aspatial data, possibly stored in a spatial database. An important application area where spatial data mining techniques can be effectively used is offered by GIS. This paper presents an extension of a prototypical GIS, named INGENS (Inductive Geographic Information System), which integrates data mining tools to support both predictive (i.e., classification) and descriptive (i.e., association discovery) tasks. The entire data mining process is devised for a user who controls the parameters of the process by means of a query written in SDMOQL, a spatial mining query language, whose design is based on the standard OQL. Five primitives have been considered as guidelines for the design of SDMOQL, namely, the set of task-relevant data to be mined, the kind of knowledge to be mined, the background knowledge to be used in the discovery process, the interestingness measures and thresholds for pattern evaluation, and the expected representation for visualizing the discovered patterns. We illustrate an application of SDMOQL to the task of Apulia map interpretation in order to discover classification rules of interest to town planners.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.