Keywords
human-altered streams, ecological status, nutrient retention, artificial neural networks, orthogonal search-based rule extraction
Start Date
1-7-2006 12:00 AM
Abstract
The recent Water Framework Directive of the European Union set year 2015 as their target for freshwater and coastal ecosystems all across Europe to achieve good ecological status. This study concerns the analysis of the empirical data from the knowledge base of an environmental decision support system developed within the European project STREAMES. These data, which come from several low-order streams located mostly on the Mediterranean region, consist of measurements of several physical, chemical and biological variables. We aim to classify these data according to the ecological status of the streams they correspond to, where stream nutrient retention efficiency (a functional ecosystem attribute) is used as an indicator of ecological status. This classification task is performed using supervised Artificial Neural Networks. The interpretability of the obtained classification results can be improved by their description in terms of simple, actionable rules. This is accomplished through the application of Orthogonal Search-based Rule Extraction, a novel overlapping rule extraction method. All the newly acquired knowledge should help water managers to focus their efforts on strategies that minimize the negative human impacts on vulnerable low-order streams.
On the prediction of the ecological status of human-altered streams and its rule-based interpretation
The recent Water Framework Directive of the European Union set year 2015 as their target for freshwater and coastal ecosystems all across Europe to achieve good ecological status. This study concerns the analysis of the empirical data from the knowledge base of an environmental decision support system developed within the European project STREAMES. These data, which come from several low-order streams located mostly on the Mediterranean region, consist of measurements of several physical, chemical and biological variables. We aim to classify these data according to the ecological status of the streams they correspond to, where stream nutrient retention efficiency (a functional ecosystem attribute) is used as an indicator of ecological status. This classification task is performed using supervised Artificial Neural Networks. The interpretability of the obtained classification results can be improved by their description in terms of simple, actionable rules. This is accomplished through the application of Orthogonal Search-based Rule Extraction, a novel overlapping rule extraction method. All the newly acquired knowledge should help water managers to focus their efforts on strategies that minimize the negative human impacts on vulnerable low-order streams.
