Learning Characteristic Rules in Geographic Information Systems

Ansaf Salleb-Aouissi (1) , Christel Vrain (2) , Daniel Cassard (3)

1 CCLS - Center for Computational Learning Systems
2 LIFO - Laboratoire d'Informatique Fondamentale d'Orléans
3 BRGM - Bureau de Recherches Géologiques et Minières

We provide a general framework for learning characterization rules of a set of objects in Geographic Information Systems (GIS) relying on the definition of distance quantified paths, and allowing flexible quantifiers. Such expressions specify how to navigate between the different layers of the GIS starting from the target setof objects to characterize. We have defined a generality relation between quantified paths and proved that it is monotonous with respect to the notion of coverage, thus allowing to develop an interactive and effective algorithm to explore the search space of possible rules. We describe GISMiner, an interactive system that we have developed based on our framework. Finally, we present our experimental results from a real GIS about mineral exploration.

Mots clés: Rule learning; Geographic Information System

Domaines: Apprentissage [cs.LG]; Informatique [cs]

Ansaf Salleb-Aouissi, Christel Vrain, Daniel Cassard. Learning Characteristic Rules in Geographic Information Systems. Rule Technologies: Foundations, Tools, and Applications, 9th International Symposium, (RuleML 2015), Aug 2015, Berlin, Germany. ⟨10.1007/978-3-319-21542-6_28⟩. ⟨hal-01182794⟩