Dissection of trained neural network hydrologic models for knowledge extraction

[1] Artificial neural networks (ANNs) are powerful tools for the modeling and forecasting of complex engineering systems and have been exploited by researchers to solve a variety of problems over the last couple of decades. In spite of their proven ability to provide superior model performance compared to traditional modeling approaches, they have not become popular among decision makers for operational use. It is probably because of their perceived black box nature that does not explain or consider the underlying physical processes involved. This paper presents the results of a study aimed at a systematic dissection of the massively parallel architectures of trained ANN hydrologic models to determine if they learn the underlying physical subprocesses during training. This has been achieved using simple qualitative and quantitative techniques. The data derived from three contrasting catchments at two different time scales were employed to develop ANN models and test the methodologies employed for knowledge extraction. The results obtained in this study indicate that the number of hidden neurons determined during training for a particular data set correspond to certain subprocesses of the overall physical process being modeled. It has been found that the time scale of the data employed has an effect on optimum ANN architecture and knowledge extracted.

[1]  K. P. Sudheer,et al.  Explaining the internal behaviour of artificial neural network river flow models , 2004 .

[2]  Yu-Chi Ho,et al.  Perturbation analysis: concepts and algorithms , 1992, WSC '92.

[3]  R. Abrahart,et al.  Detection of conceptual model rainfall—runoff processes inside an artificial neural network , 2003 .

[4]  B. S. Thandaveswara,et al.  A non-linear rainfall–runoff model using an artificial neural network , 1999 .

[5]  Kuolin Hsu,et al.  Artificial Neural Network Modeling of the Rainfall‐Runoff Process , 1995 .

[6]  K. P. Sudheer,et al.  Knowledge Extraction from Trained Neural Network River Flow Models , 2005 .

[7]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[8]  Ashu Jain,et al.  Short-Term Water Demand Forecast Modelling at IIT Kanpur Using Artificial Neural Networks , 2001 .

[9]  K. P. Sudheer,et al.  Identification of physical processes inherent in artificial neural network rainfall runoff models , 2004 .

[10]  Rao S. Govindaraju,et al.  Prediction of watershed runoff using Bayesian concepts and modular neural networks , 2000 .

[11]  Ashu Jain,et al.  Development of effective and efficient rainfall‐runoff models using integration of deterministic, real‐coded genetic algorithms and artificial neural network techniques , 2004 .

[12]  José Manuel Benítez,et al.  Interpretation of artificial neural networks by means of fuzzy rules , 2002, IEEE Trans. Neural Networks.

[13]  Ignacio Requena,et al.  Are artificial neural networks black boxes? , 1997, IEEE Trans. Neural Networks.

[14]  Ashu Jain,et al.  Visualisation of Hidden Neuron Behaviour in a Neural Network Rainfall-Runoff Model , 2009 .

[15]  Ashu Jain,et al.  Comparative Analysis of Event-Based Rainfall-Runoff Modeling Techniques—Deterministic, Statistical, and Artificial Neural Networks , 2003 .

[16]  Momcilo Markus,et al.  PRECIPITATION-RUNOFF MODELING USING ARTIFICIAL NEURAL NETWORKS AND CONCEPTUAL MODELS , 2000 .