A class of hybrid morphological perceptrons with application in time series forecasting

In this work a class of hybrid morphological perceptrons, called dilation-erosion perceptron (DEP), is presented to overcome the random walk dilemma in the time series forecasting problem. It consists of a convex combination of fundamental operators from mathematical morphology (MM) on complete lattices theory (CLT). A gradient steepest descent method is presented to design the proposed DEP (learning process), using the back propagation (BP) algorithm and a systematic approach to overcome the problem of nondifferentiability of morphological operators. The learning process includes an automatic phase fix procedure that is geared at eliminating time phase distortions observed in some time series. Finally, an experimental analysis is conducted with the proposed DEP using five real world time series, where five well-known performance metrics and an evaluation function are used to assess the forecasting performance of the proposed model. The obtained results are compared with those generated by classical forecasting models presented in the literature.

[1]  G. Grätzer,et al.  Lattice Theory: First Concepts and Distributive Lattices , 1971 .

[2]  Petros Maragos.  A Representation Theory for Morphological Image and Signal Processing , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Carsten Peterson,et al.  Finding the Embedding Dimension and Variable Dependencies in Time Series , 1994, Neural Computation.

[4]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[5]  A. Walden,et al.  Spectral analysis for physical applications : multitaper and conventional univariate techniques , 1996 .

[6]  Johan A. K. Suykens,et al.  Financial time series prediction using least squares support vector machines within the evidence framework , 2001, IEEE Trans. Neural Networks.

[7]  Tiago Alessandro Espínola Ferreira,et al.  Morphological-Rank-Linear Time-lag Added Evolutionary Forecasting method for financial time series forecasting , 2008, 2008 IEEE Congress on Evolutionary Computation (IEEE World Congress on Computational Intelligence).

[8]  Aydogan Savran,et al.  Multifeedback-Layer Neural Network , 2007, IEEE Transactions on Neural Networks.

[9]  R. Savit,et al.  Time series and dependent variables , 1991 .

[10]  Oscar Castillo,et al.  Hybrid intelligent systems for time series prediction using neural networks, fuzzy logic, and fractal theory , 2002, IEEE Trans. Neural Networks.

[11]  Vassilios Petridis,et al.  Fuzzy Lattice Neurocomputing (FLN) models , 2000, Neural Networks.

[12]  Tiago Alessandro Espínola Ferreira,et al.  An evolutionary Morphological-Rank-Linear approach for time series prediction , 2007, 2007 IEEE Congress on Evolutionary Computation.

[13]  Tiago Alessandro Espínola Ferreira,et al.  An intelligent hybrid morphological-rank-linear method for financial time series prediction , 2009, Neurocomputing.

[14]  Jennifer L. Davidson,et al.  Morphology neural networks: An introduction with applications , 1993 .

[15]  Peter Sussner,et al.  Morphological and Certain Fuzzy Morphological Associative Memories for Classification and Prediction , 2007, Computational Intelligence Based on Lattice Theory.

[16]  Henry Leung,et al.  Prediction of noisy chaotic time series using an optimal radial basis function neural network , 2001, IEEE Trans. Neural Networks.

[17]  Guoqiang Peter Zhang,et al.  A neural network ensemble method with jittered training data for time series forecasting , 2007, Inf. Sci..

[18]  Manuel Graña,et al.  On the Application of Associative Morphological Memories to Hyperspectral Image Analysis , 2003, IWANN.

[19]  Christian Ronse,et al.  Why mathematical morphology needs complete lattices , 1990, Signal Process..

[20]  F. Takens.  Detecting strange attractors in turbulence , 1981 .

[21]  Renate Sitte,et al.  Neural Networks Approach to the Random Walk Dilemma of Financial Time Series , 2002, Applied Intelligence.

[22]  Peter Sussner,et al.  Constructive Morphological Neural Networks: Some Theoretical Aspects and Experimental Results in Classification , 2009, Constructive Neural Networks.

[23]  Ali Rahimi.  Learning to Transform Time Series with a Few Examples , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[24]  J. N. Wilson,et al.  Image Algebra: An Overview , 1990, Comput. Vis. Graph. Image Process..

[25]  Vicenç Torra,et al.  Towards the evaluation of time series protection methods , 2009, Inf. Sci..

[26]  Vassilios Petridis,et al.  Fuzzy lattice neural network (FLNN): a hybrid model for learning , 1998, IEEE Trans. Neural Networks.

[27]  Tiago Alessandro Espínola Ferreira,et al.  A New Intelligent System Methodology for Time Series Forecasting with Artificial Neural Networks , 2008, Neural Processing Letters.

[28]  María Cristina Riff,et al.  Towards an immune system that solves CSP , 2007, 2007 IEEE Congress on Evolutionary Computation.

[29]  Vladimir Pavlovic,et al.  Discriminative Learning for Dynamic State Prediction , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[30]  Paul D. Gader,et al.  LADAR target detection using morphological shared-weight neural networks , 2000, Machine Vision and Applications.

[31]  Ricardo de A. Araújo.  Swarm-Based Hybrid Intelligent Forecasting Method For Financial Time Series Prediction , 2007 .

[32]  D. Li,et al.  Image Algebra And Its Relationship To Neural Networks , 1989, Defense, Security, and Sensing.

[33]  Godfrey C. Onwubolu,et al.  Design of hybrid differential evolution and group method of data handling networks for modeling and prediction , 2008, Inf. Sci..

[34]  Ricardo de A. Araújo,et al.  Hybrid intelligent methodology to design translation invariant morphological operators for Brazilian stock market prediction , 2010, Neural Networks.

[35]  Gerhard X. Ritter,et al.  Theory of morphological neural networks , 1990, Photonics West - Lasers and Applications in Science and Engineering.

[36]  Ricardo de A. Araújo.  Swarm-based translation-invariant morphological prediction method for financial time series forecasting , 2010, Inf. Sci..

[37]  Simon Haykin,et al.  Neural Networks and Learning Machines , 2010 .

[38]  Tiago Alessandro Espínola Ferreira,et al.  An Intelligent Hybrid Approach for Designing Increasing Translation Invariant Morphological Operators for Time Series Forecasting , 2007, ISNN.

[39]  Paul D. Gader,et al.  Domain learning using Choquet integral-based morphological shared weight neural networks , 2003, Image Vis. Comput..

[40]  Peter Sussner,et al.  Implicative Fuzzy Associative Memories , 2006, IEEE Transactions on Fuzzy Systems.

[41]  H. Kantz,et al.  Nonlinear time series analysis , 1997 .

[42]  Tiago Alessandro Espínola Ferreira,et al.  Improved Evolutionary Hybrid Method for Designing Morphological Operators , 2006, 2006 International Conference on Image Processing.

[43]  Ricardo de A. Araújo.  A Hybrid Intelligent Morphological Approach for Stock Market Forecasting , 2010, Neural Processing Letters.

[44]  Hak-Keung Lam,et al.  Tuning of the structure and parameters of a neural network using an improved genetic algorithm , 2003, IEEE Trans. Neural Networks.

[45]  H. Heijmans.  Morphological image operators , 1994 .

[46]  Coskun Hamzaçebi,et al.  Improving artificial neural networks' performance in seasonal time series forecasting , 2008, Inf. Sci..

[47]  Ricardo de A. Araújo,et al.  Modular Morphological Neural Network Training via Adaptive Genetic Algorithm for Designing Translation Invariant Operators , 2006, 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.

[48]  Peter Sussner.  Kernels for Morphological Associative Memories , 1998, NC.

[49]  Jean Serra,et al.  Image Analysis and Mathematical Morphology , 1983 .

[50]  P. Young,et al.  Time series analysis, forecasting and control , 1972, IEEE Transactions on Automatic Control.

[51]  Peter Sussner,et al.  Morphological associative memories , 1998, IEEE Trans. Neural Networks.

[52]  Petros Maragos,et al.  Neural networks with hybrid morphological/rank/linear nodes and their application to handwritten character recognition , 1998, 9th European Signal Processing Conference (EUSIPCO 1998).

[53]  Gerhard X. Ritter,et al.  Lattice algebra approach to single-neuron computation , 2003, IEEE Trans. Neural Networks.

[54]  H. Kunzi,et al.  Lectu re Notes in Economics and Mathematical Systems , 1975 .

[55]  Joseph N. Wilson,et al.  Handbook of computer vision algorithms in image algebra , 1996 .

[56]  N. Tanaka,et al.  Estimating the active demension of the dynamics in a time series based on an information criterion , 2001 .

[57]  Peter Sussner,et al.  Gray-scale morphological associative memories , 2006, IEEE Transactions on Neural Networks.

[58]  Junior Barrera,et al.  Decomposition of mappings between complete lattices by mathematical morphology, Part I. General lattices , 1993, Signal Process..

[59]  Min Han,et al.  Support Vector Echo-State Machine for Chaotic Time-Series Prediction , 2007, IEEE Transactions on Neural Networks.

[60]  Peter Sussner,et al.  A general framework for fuzzy morphological associative memories , 2008, Fuzzy Sets Syst..

[61]  Bogdan Raducanu,et al.  Morphological Scale Spaces and Associative Morphological Memories: Results on Robustness and Practical Applications , 2004, Journal of Mathematical Imaging and Vision.

[62]  Paul D. Gader,et al.  Automatic target detection using entropy optimized shared-weight neural networks , 2000, IEEE Trans. Neural Networks Learn. Syst..

[63]  Michael P. Clements,et al.  On the limitations of comparing mean square forecast errors , 1993 .

[64]  Paul D. Gader,et al.  Morphological regularization neural networks , 2000, Pattern Recognit..

[65]  Babak Nadjar Araabi,et al.  Predicting Chaotic Time Series Using Neural and Neurofuzzy Models: A Comparative Study , 2006, Neural Processing Letters.

[66]  Francisco Marcos de Assis,et al.  Designing translation invariant operations via neural network training , 2000, Proceedings 2000 International Conference on Image Processing (Cat. No.00CH37101).

[67]  Peter Sussner,et al.  An introduction to morphological perceptrons with competitive learning , 2009, 2009 International Joint Conference on Neural Networks.

[68]  Jennifer L. Davidson,et al.  Template learning in morphological neural nets , 1991, Optics & Photonics.

[69]  Peter Sussner,et al.  An introduction to morphological neural networks , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[70]  Tiago A. E. Ferreira,et al.  An Evolutionary Morphological Approach for Financial Time Series Forecasting , 2006, 2006 IEEE International Conference on Evolutionary Computation.

[71]  Gerhard X. Ritter,et al.  Computational Intelligence Based on Lattice Theory , 2007, Studies in Computational Intelligence.

[72]  Peter Sussner,et al.  Morphological perceptrons with competitive learning: Lattice-theoretical framework and constructive learning algorithm , 2011, Inf. Sci..

[73]  Petros Maragos,et al.  Morphological/rank neural networks and their adaptive optimal design for image processing , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.

[74]  Ricardo de A. Araújo.  Translation Invariant Morphological Time-lag Added Evolutionary Forecasting method for stock market prediction , 2011, Expert Syst. Appl..