A neural-fuzzy approach to classify the ecological status in surface waters
Autor: | Marta Schuhmacher, William Ocampo-Duque, José L. Domingo |
---|---|
Rok vydání: | 2007 |
Předmět: |
Artificial neural network
Computer science Ecology Health Toxicology and Mutagenesis Water Pollution Uncertainty Probabilistic logic Sampling (statistics) General Medicine Fuzzy control system Toxicology Pollution Causality Data set Statistical classification Neural fuzzy Fuzzy Logic Rivers Water Framework Directive Neural Networks Computer Algorithms Ecosystem |
Zdroj: | Environmental Pollution. 148:634-641 |
ISSN: | 0269-7491 |
Popis: | A methodology based on a hybrid approach that combines fuzzy inference systems and artificial neural networks has been used to classify ecological status in surface waters. This methodology has been proposed to deal efficiently with the non-linearity and highly subjective nature of variables involved in this serious problem. Ecological status has been assessed with biological, hydro-morphological, and physicochemical indicators. A data set collected from 378 sampling sites in the Ebro river basin has been used to train and validate the hybrid model. Up to 97.6% of sampling sites have been correctly classified with neural-fuzzy models. Such performance resulted very competitive when compared with other classification algorithms. With non-parametric classification-regression trees and probabilistic neural networks, the predictive capacities were 90.7% and 97.0%, respectively. The proposed methodology can support decision-makers in evaluation and classification of ecological status, as required by the EU Water Framework Directive. |
Databáze: | OpenAIRE |
Externí odkaz: |