Optimization of the ANNs Predictive Capability Using the Taguchi Approach: A Case Study

Artificial neural networks (ANNs) are a valid alternative predictive method to the traditional statistical techniques currently used in many research fields where a massive amount of data is challenging to manage. In environmental analysis, ANNs can analyze pollution sources in large areas, estimating difficult and expensive to detect contaminants from other easily measurable pollutants, especially for screening procedures. In this study, organic micropollutants have been predicted from heavy metals concentration using ANNs. Sampling was performed in an agricultural field where organic and inorganic contaminants concentrations are beyond the legal limits. A critical problem of a neural network design is to select its parametric topology, which can prejudice the reliability of the model. Therefore, it is very important to assess the performance of ANNs when applying different types of parameters of the net. In this work, based on Taguchi L12 orthogonal array, turning experiments were conducted to identify the best parametric set of an ANNs design, considering different combinations of sample number, scaling, training rate, activation functions, number of hidden layers, and epochs. The composite desirability value for the multi-response variables has been obtained through the desirability function analysis (DFA). The parameters’ optimum levels have been identified using this methodology.

[1]  Maria Grazia Bonelli,et al.  Artificial neural networks to evaluate organic and inorganic contamination in agricultural soils. , 2017, Chemosphere.

[2]  Pengjiang Qian,et al.  Cat Swarm Optimization applied to alcohol use disorder identification , 2018, Multimedia Tools and Applications.

[3]  Michael I. Jordan,et al.  Learning with Mixtures of Trees , 2001, J. Mach. Learn. Res..

[4]  Jasmina Arifovic,et al.  Using genetic algorithms to select architecture of a feedforward artificial neural network , 2001 .

[5]  George-Christopher Vosniakos,et al.  Optimizing feedforward artificial neural network architecture , 2007, Eng. Appl. Artif. Intell..

[6]  Yoshua Bengio,et al.  Exploring Strategies for Training Deep Neural Networks , 2009, J. Mach. Learn. Res..

[8]  Holger R. Maier,et al.  The effect of internal parameters and geometry on the performance of back-propagation neural networks: an empirical study , 1998 .

[9]  Kurt Hornik,et al.  Approximation capabilities of multilayer feedforward networks , 1991, Neural Networks.

[10]  Michael Y. Hu,et al.  Effect of data standardization on neural network training , 1996 .

[11]  Jürgen Schmidhuber,et al.  Deep learning in neural networks: An overview , 2014, Neural Networks.

[12]  Cristian Caizar,et al.  Application of Taguchi method to selection of optimal lubrication and cutting conditions in face milling of AlMg3 , 2011 .

[13]  Zaher Mundher Yaseen,et al.  ANN Based Sediment Prediction Model Utilizing Different Input Scenarios , 2015, Water Resources Management.

[14]  Jiju Antony,et al.  Multi‐response optimization in industrial experiments using Taguchi's quality loss function and principal component analysis , 2000 .

[15]  John L. Yang.  A Systematic Approach for Identifying Optimum Surface Roughness Performance in End-Milling Operations By Mr . , 2001 .

[16]  Muammer Nalbant,et al.  Application of Taguchi method in the optimization of cutting parameters for surface roughness in turning , 2007 .

[17]  W S McCulloch,et al.  A logical calculus of the ideas immanent in nervous activity , 1990, The Philosophy of Artificial Intelligence.

[18]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[19]  Francisco Venegas-Martínez,et al.  A Review of Artificial Neural Networks: How Well Do They Perform in Forecasting Time Series? , 2013 .

[20]  S. Agatonovic-Kustrin,et al.  Basic concepts of artificial neural network (ANN) modeling and its application in pharmaceutical research. , 2000, Journal of pharmaceutical and biomedical analysis.

[21]  R. Erb,et al.  Introduction to Backpropagation Neural Network Computation , 1993, Pharmaceutical Research.

[22]  Yasuaki Kuroe,et al.  On Activation Functions for Complex-Valued Neural Networks - Existence of Energy Functions , 2003, ICANN.

[23]  Michael Georgiopoulos,et al.  Coupling weight elimination with genetic algorithms to reduce network size and preserve generalization , 1997, Neurocomputing.

[24]  A. Ghaffari,et al.  Performance comparison of neural network training algorithms in modeling of bimodal drug delivery. , 2006, International journal of pharmaceutics.

[25]  Gordon Leitch,et al.  Economic Forecast Evaluation: Profits versus the Conventional Error Measures , 1991 .

[26]  Xiaosheng Wu,et al.  Diagnosis of COVID-19 by Wavelet Renyi Entropy and Three-Segment Biogeography-Based Optimization , 2020, Int. J. Comput. Intell. Syst..

[27]  N. K. Sharma,et al.  Quality Loss Function - Common Methodology for Nominal-The-Best, Smaller-The-Better, and Larger-The-Better Cases , 2007 .

[28]  Khashayar Khorasani,et al.  A new strategy for adaptively constructing multilayer feedforward neural networks , 2003, Neurocomputing.

[29]  Farouq S Mjalli,et al.  Use of artificial neural network black-box modeling for the prediction of wastewater treatment plants performance. , 2007, Journal of environmental management.

[30]  Dinesh Kumar,et al.  Optimization of neural network parameters using Grey–Taguchi methodology for manufacturing process applications , 2015 .

[31]  Alice E. Smith,et al.  Performance measures, consistency, and power for artificial neural network models , 1995 .

[32]  Holger R. Maier,et al.  Understanding the behaviour and optimising the performance of back-propagation neural networks: an empirical study , 1998 .

[33]  Mujahid Tabassum,et al.  A GENETIC ALGORITHM ANALYSIS TOWARDS OPTIMIZATION SOLUTIONS , 2014 .

[34]  M. W Gardner,et al.  Artificial neural networks (the multilayer perceptron)—a review of applications in the atmospheric sciences , 1998 .

[35]  Holger R. Maier,et al.  DATA DIVISION FOR DEVELOPING NEURAL NETWORKS APPLIED TO GEOTECHNICAL ENGINEERING , 2004 .

[36]  Cheng-Jian Lin,et al.  Using Convolutional Neural Networks Based on a Taguchi Method for Face Gender Recognition , 2020 .

[37]  Archontoula Chaloulakou,et al.  Comparative assessment of neural networks and regression models for forecasting summertime ozone in Athens. , 2003, The Science of the total environment.

[38]  A. V. Olgac,et al.  Performance Analysis of Various Activation Functions in Generalized MLP Architectures of Neural Networks , 2011 .

[39]  D.R. Hush,et al.  Progress in supervised neural networks , 1993, IEEE Signal Processing Magazine.

[40]  A. Bonfitto,et al.  Estimation Accuracy and Computational Cost Analysis of Artificial Neural Networks for State of Charge Estimation in Lithium Batteries , 2019, Batteries.

[41]  C George,et al.  A Balancing Act: Optimizing a Product's Properties , 1994 .

[42]  Francesco Musumeci,et al.  A Survey on Application of Machine Learning Techniques in Optical Networks , 2018, ArXiv.

[43]  Anil K. Jain,et al.  Artificial Neural Networks: A Tutorial , 1996, Computer.

[44]  Byung Ro Moon,et al.  A Hybrid Neurogenetic Approach for Stock Forecasting , 2007, IEEE Transactions on Neural Networks.

[45]  Boby John.  Application of desirability function for optimizing the performance characteristics of carbonitrided bushes , 2013 .