Mining a database on Alsatian rivers

Corinne Grac (1) , Agnès Herrmann (1) , Florence Le Ber (1, 2) , Michèle Trémolières (1) , Agnès Braud (3) , Adamou Handja (3) , Nicolas Lachiche (3)

1 CEVH - Centre d'Ecologie Végétale et d'Hydrologie
2 ORPAILLEUR - Knowledge representation, reasonning
3 LSIIT - Laboratoire des Sciences de l'Image, de l'Informatique et de la Télédétection

We aim at comparing the answers of the bio-indication tools of the water bodies towards the various pressures which they undergo. Therefore we have designed of a database gathering the physical, physico-chemical, floristic (diatoms and macrophytes), and faunistic data (invertebrates, oligochaetes, fishes) of the rivers streams and water areas of the Alsace plain. Besides, we are implementing data mining methods to explore the database. These methods are used to find out regularities and similarities that can be interpreted by a domain expert. Indeed, we consider data mining as a part of a more general process that is knowledge discovery in databases. We give examples of such methods and the results that can be obtained.

Mots clés: data mining; water quality; bio-indication tools

Domaines: Intelligence artificielle [cs.AI]; Interfaces continentales, environnement

Corinne Grac, Agnès Herrmann, Florence Le Ber, Michèle Trémolières, Agnès Braud, et al.. Mining a database on Alsatian rivers. 7th International Conference on Hydroinformatics, 2006, France. pp.2263-2270. ⟨hal-00107053⟩