In recent times, several extensions of data mining methods and techniques have been explored aiming at dealing with advanced databases. Many promising applications of inductive logic programming (ILP) to knowledge discovery in databases have also emerged in order to benefit from semantics and inference rules of first-order logic. In this paper, an ILP framework for frequent pattern discovery in spatial data is presented. The pattern discovery algorithm operates on first-order logic descriptions computed by an initial step of feature extraction from a spatial database. The algorithm benefits of the available background knowledge on the spatial domain and systematically explores the hierarchical structure of task-relevant geographic layers. Preliminary results have been obtained by running the algorithm SPADA on spatial data from an Italian province.
A LOGICAL FRAMEWORK FOR FREQUENT PATTERN DISCOVERY IN SPATIAL DATA
MALERBA, Donato;ESPOSITO, Floriana;LISI, Francesca Alessandra
2001-01-01
Abstract
In recent times, several extensions of data mining methods and techniques have been explored aiming at dealing with advanced databases. Many promising applications of inductive logic programming (ILP) to knowledge discovery in databases have also emerged in order to benefit from semantics and inference rules of first-order logic. In this paper, an ILP framework for frequent pattern discovery in spatial data is presented. The pattern discovery algorithm operates on first-order logic descriptions computed by an initial step of feature extraction from a spatial database. The algorithm benefits of the available background knowledge on the spatial domain and systematically explores the hierarchical structure of task-relevant geographic layers. Preliminary results have been obtained by running the algorithm SPADA on spatial data from an Italian province.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
