The Use of the Galois lattice for the extraction and the visualization of the spatial association rules
Rabia Marghoubi, A. Boulmakoul, Karine Zeitouni · 2006
Spatial data mining is a challenging issue in data mining research area. In this paper we present a novel algorithm for mining spatial association rules. Those rules are meant to discover hidden spatial relationships between located phenomena. Our approach is divided in two steps. The first one elaborates a spatial context, while the second step focuses in pattern mining. To this end, we use the Galois lattice paradigm. Indeed, this lattice allows efficient mining of association rules based on closed itemsets. This results in an effective and efficient method for mining spatial association rules. Resides, we implement the visualisation of extracted spatial association rules by extending Galicia platform to spatial context. This paper underpins our approach, the prototype and provides the experimental results. Our target application is related to the promotion of the use of Internet in Morocco, and more precisely the improvement of the added value services