Input Selection of Wavelet-Coupled Neural Network Models for Rainfall-Runoff Modelling

2018 ◽  
Vol 33 (3) ◽  
pp. 955-973 ◽  
Author(s):  
Muhammad Shoaib ◽  
Asaad Y. Shamseldin ◽  
Sher Khan ◽  
Muhammad Sultan ◽  
Fiaz Ahmad ◽  
...  
Keyword(s):  
Neural Network ◽  
Network Models ◽  
Rainfall Runoff ◽  
Neural Network Models ◽  
Input Selection ◽  
Selection Of

scholarly journals Neural network emulation of a rainfall-runoff model

2007 ◽  
Vol 4 (1) ◽  
pp. 287-326 ◽  
Author(s):  
R. J. Abrahart ◽  
L. M. See
Keyword(s):  
Neural Network ◽  
Transfer Functions ◽  
Network Models ◽  
Exact Equation ◽  
Rainfall Runoff ◽  
Neural Network Models ◽  
Linear Relationships ◽  
Non Linear ◽  
Rainfall Runoff Model ◽  
Runoff Model

Abstract. The potential of an artificial neural network to perform simple non-linear hydrological transformations is examined. Four neural network models were developed to emulate different facets of a recognised non-linear hydrological transformation equation that possessed a small number of variables and contained no temporal component. The modeling process was based on a set of uniform random distributions. The cloning operation facilitated a direct comparison with the exact equation-based relationship. It also provided broader information about the power of a neural network to emulate existing equations and model non-linear relationships. Several comparisons with least squares multiple linear regression were performed. The first experiment involved a direct emulation of the Xinanjiang Rainfall-Runoff Model. The next two experiments were designed to assess the competencies of two neural solutions that were developed on a reduced number of inputs. This involved the omission and conflation of previous inputs. The final experiment used derived variables to model intrinsic but otherwise concealed internal relationships that are of hydrological interest. Two recent studies have suggested that neural solutions offer no worthwhile improvements in comparison to traditional weighted linear transfer functions for capturing the non-linear nature of hydrological relationships. Yet such fundamental properties are intrinsic aspects of catchment processes that cannot be excluded or ignored. The results from the four experiments that are reported in this paper are used to challenge the interpretations from these two earlier studies and thus further the debate with regards to the appropriateness of neural networks for hydrological modelling.

scholarly journals Classification of heterogeneous precipitation fields for the assessment and possible improvement of lumped neural network models for streamflow forecasts

2006 ◽  
Vol 3 (1) ◽  
pp. 201-227 ◽  
Author(s):  
N. Lauzon ◽  
F. Anctil ◽  
C. W. Baxter
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Multilayer Perceptron ◽  
Network Models ◽  
Rainfall Runoff ◽  
Neural Network Models ◽  
Kohonen Neural Network ◽  
Streamflow Forecasts ◽  
Rainfall Runoff Model

Abstract. This work addresses the issue of better considering the heterogeneity of precipitation fields within lumped rainfall-runoff models where only areal mean precipitation is usually used as an input. A method using a Kohonen neural network is proposed for the classification of precipitation fields. The evaluation and improvement of the performance of a lumped rainfall-runoff model for one-day ahead predictions is then established based on this classification. Multilayer perceptron neural networks are employed as lumped rainfall-runoff models. The Bas-en-Basset watershed in France, which is equipped with 23 rain gauges with data for a 21-year period, is employed as the application case. The results demonstrate the relevance of the proposed classification method, which produces groups of precipitation fields that are in agreement with the global climatological features affecting the region, as well as with the topographic constraints of the watershed (i.e., orography). The strengths and weaknesses of the rainfall-runoff models are highlighted by the analysis of their performance vis-à-vis the classification of precipitation fields. The results also show the capability of multilayer perceptron neural networks to account for the heterogeneity of precipitation, even when built as lumped rainfall-runoff models.

scholarly journals Automatic migraine classification using artificial neural networks

F1000Research ◽  
2020 ◽  
Vol 9 ◽  
pp. 618
Author(s):  
Paola A. Sanchez-Sanchez ◽  
José Rafael García-González ◽  
Juan Manuel Rúa Ascar
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Network ◽  
Artificial Neural Networks ◽  
Network Models ◽  
Neural Network Models ◽  
Accuracy And Precision ◽  
Artificial Neural ◽  
Artificial Neural Network Models ◽  
Selection Of

Background: Previous studies of migraine classification have focused on the analysis of brain waves, leading to the development of complex tests that are not accessible to the majority of the population. In the early stages of this pathology, patients tend to go to the emergency services or outpatient department, where timely identification largely depends on the expertise of the physician and continuous monitoring of the patient. However, owing to the lack of time to make a proper diagnosis or the inexperience of the physician, migraines are often misdiagnosed either because they are wrongly classified or because the disease severity is underestimated or disparaged. Both cases can lead to inappropriate, unnecessary, or imprecise therapies, which can result in damage to patients’ health. Methods: This study focuses on designing and testing an early classification system capable of distinguishing between seven types of migraines based on the patient’s symptoms. The methodology proposed comprises four steps: data collection based on symptoms and diagnosis by the treating physician, selection of the most relevant variables, use of artificial neural network models for automatic classification, and selection of the best model based on the accuracy and precision of the diagnosis. Results: The artificial neural network models used provide an excellent classification performance, with accuracy and precision levels >97% and which exceed the classifications made using other model, such as logistic regression, support vector machines, nearest neighbor, and decision trees. Conclusions: The implementation of migraine classification through artificial neural networks is a powerful tool that reduces the time to obtain accurate, reliable, and timely clinical diagnoses.

scholarly journals Automatic migraine classification using artificial neural networks

F1000Research ◽  
2020 ◽  
Vol 9 ◽  
pp. 618
Author(s):  
Paola A. Sanchez-Sanchez ◽  
José Rafael García-González ◽  
Juan Manuel Rúa Ascar
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Nearest Neighbor ◽  
Emergency Services ◽  
Network Models ◽  
Support Vector ◽  
Neural Network Models ◽  
Accuracy And Precision ◽  
Artificial Neural ◽  
Selection Of

Background: Previous studies of migraine classification have focused on the analysis of brain waves, leading to the development of complex tests that are not accessible to the majority of the population. In the early stages of this pathology, patients tend to go to the emergency services or outpatient department, where timely identification largely depends on the expertise of the physician and continuous monitoring of the patient. However, owing to the lack of time to make a proper diagnosis or the inexperience of the physician, migraines are often misdiagnosed either because they are wrongly classified or because the disease severity is underestimated or disparaged. Both cases can lead to inappropriate, unnecessary, or imprecise therapies, which can result in damage to patients’ health. Methods: This study focuses on designing and testing an early classification system capable of distinguishing between seven types of migraines based on the patient’s symptoms. The methodology proposed comprises four steps: data collection based on symptoms and diagnosis by the treating physician, selection of the most relevant variables, use of artificial neural network models for automatic classification, and selection of the best model based on the accuracy and precision of the diagnosis. Results: The neural network models used provide an excellent classification performance, with accuracy and precision levels >97% and which exceed the classifications made using other model, such as logistic regression, support vector machines, nearest neighbor, and decision trees. Conclusions: The implementation of migraine classification through neural networks is a powerful tool that reduces the time to obtain accurate, reliable, and timely clinical diagnoses.

Sign in / Sign up

Export Citation Format

Share Document