We present a new Geographical Information System (GIS)-based framework encapsulating a Mamdani fuzzy rule-based system applied as classifier to partition an urban system in homogenous urban areas, called urban contexts. The Mamdani fuzzy rule system is applied to classify the census zone of the urban system based on their characteristics we dissolve adjoining census zones belonging to the same urban class to form urban contexts. An evaluation of the reliability of the classification results is obtained. We test our framework on the Municipality of Pozzuoli (Italy); the results obtained with that our framework are comparable with ones inferred by domain experts.
Mamdani fuzzy rule-based system classification to partition urban systems / Cardone, Barbara; DI MARTINO, Ferdinando. - (2019), pp. 1-12. (Intervento presentato al convegno USB Proceedings The 16th International Conference on Modeling Decisions for Artificial Intelligence MDAI 2019, Milan, Italy 4-6 September 2019).
Mamdani fuzzy rule-based system classification to partition urban systems
barbara cardone;ferdinando di martino
2019
Abstract
We present a new Geographical Information System (GIS)-based framework encapsulating a Mamdani fuzzy rule-based system applied as classifier to partition an urban system in homogenous urban areas, called urban contexts. The Mamdani fuzzy rule system is applied to classify the census zone of the urban system based on their characteristics we dissolve adjoining census zones belonging to the same urban class to form urban contexts. An evaluation of the reliability of the classification results is obtained. We test our framework on the Municipality of Pozzuoli (Italy); the results obtained with that our framework are comparable with ones inferred by domain experts.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
