Showing 1 to 56 of 56 matching Articles
Results per page:
Export (CSV)
By
Gaxiola, Fernando; Melin, Patricia; Valdez, Fevrier; Castillo, Oscar
Show all (4)
1 Citations
This paper presents a new modular neural network architecture that is used to build a system for pattern recognition based on the iris biometric measurement of persons. In this system, the properties of the person iris database are enhanced with image processing methods, and the coordinates of the center and radius of the iris are obtained to make a cut of the area of interest by removing the noise around the iris. The inputs to the modular neural network are the processed iris images and the output is the number of the identified person. The integration of the modules was done with a type2 fuzzy integrator at the level of the sub modules, and with a gating network at the level of the modules.
more …
By
Sánchez, Daniela; Melin, Patricia; Castillo, Oscar
In this paper we propose a new model of a Modular Neural Network (MNN) with fuzzy integration based on granular computing. The topology and parameters of the model are optimized with a Hierarchical Genetic Algorithm (HGA). The model was applied to the case of human recognition to illustrate its applicability. The proposed method is able to divide the data automatically into sub modules, to work with a percentage of images and select which images will be used for training. We considered, to test this method, the problem of human recognition based on ear, and we used a database with 77 persons (with 4 images each person for this task).
more …
By
Melin, Patricia; Ochoa, Valente; Valenzuela, Luis; Torres, Gabriela; Clemente, Daniel
Show all (5)
Abstract
We describe in this paper the application of several neural network architectures to the problem of simulating and predicting the dynamic behavior of complex economic time series. We use several neural network models and training algorithms to compare the results and decide at the end, which one is best for this application. We also compare the simulation results with the traditional approach of using a statistical model. In this case, we use real time series of prices of consumer goods to test our models. Real prices of tomato and green onion in the U.S. show complex fluctuations in time and are very complicated to predict with traditional statistical approaches.
more …
By
Melin, Patricia; Gonzalez, Claudia; Bravo, Diana; Gonzalez, Felma; Martinez, Gabriela
Show all (5)
8 Citations
Abstract
We describe in this paper a new approach for pattern recognition using modular neural networks with a fuzzy logic method for response integration. We proposed a new architecture for modular neural networks for achieving pattern recognition in the particular case of human faces and fingerprints. Also, the method for achieving response integration is based on the fuzzy Sugeno integral with some modifications. Response integration is required to combine the outputs of all the modules in the modular network. We have applied the new approach for fingerprint and face recognition with a real database from students of our institution.
more …
By
Gaxiola, Fernando; Melin, Patricia; Valdez, Fevrier; Castillo, Oscar
Show all (4)
2 Citations
In this paper a neural network learning method with type2 fuzzy weight adjustment is proposed. The mathematical analysis of the proposed learning method architecture and the adaptation of type2 fuzzy weights are presented. The proposed method is based on research of recent methods that handle weight adaptation and especially the use of fuzzy weights. In this work an ensemble neural network of three neural networks and the use of average integration to obtain the final result is presented. The proposed approach is applied to a case of time series prediction to illustrate the advantage of using type2 fuzzy weights.
more …
By
Sánchez, Daniela; Melin, Patricia; Castillo, Oscar
In this paper, a new method for fuzzy inference system optimization is proposed. The optimization consists in find the optimal parameters of fuzzy inference system used to combine the responses of modular neural networks using a hierarchical genetic algorithm. The optimized parameters are: type of fuzzy logic (type1 and interval type2), type of system (Mamdani or Sugeno), type of membership functions, number of membership functions in each variable (inputs and output), their parameters and the consequents of the fuzzy rules. Four benchmark databases are used to test the proposed method where, each database is a different biometric measure (face, iris, ear and voice) and each database is learned by a modular neural network. The main objective of the fuzzy inference system is to combine the different responses of the modular neural network and achieve final good results even when one (o more) biometric measure has individually a bad result. The results obtained in a previous work are used to compare with the results obtained in this paper.
more …
By
Hernández, Paula Hernández; CruzReyes, Laura; Melin, Patricia; MarOrtiz, Julio; Fraire Huacuja, Héctor Joaquín; Soberanes, Héctor José Puga; Barbosa, Juan Javier González
Show all (7)
1 Citations
This paper approaches the containership stowage problem. It is an NPhard minimization problem whose goal is to find optimal plans for stowing containers into a containership with low operational costs, subject to a set of structural and operational constraints. In this work, we apply to this problem an antbased hyperheuristic algorithm for the first time, according to our literature review. Ant colony and hyperheuristic algorithms have been successfully used in others application domains. We start from the initial solution, based in relaxed ILP model; then, we look for the global ship stability of the overall stowage plan by using a hyperheuristic approach. Besides, we reduce the handling time of the containers to be loaded on the ship. The validation of the proposed approach is performed by solving some pseudorandomly generated instances constructed through ranges based in reallife values obtained from the literature.
more …
By
Sánchez, Daniela; Melin, Patricia; Castillo, Oscar; Valdez, Fevrier
Show all (4)
1 Citations
In this paper a new model of a Modular Neural Network (MNN) with fuzzy integration based on granular computing is proposed. The topology and parameters of the MNN are optimized with a Hierarchical Genetic Algorithm (HGA). The proposed method can divide the data automatically into sub modules or granules, chooses the percentage of images and selects which images will be used for training. The responses of each sub module are combined using a fuzzy integrator, the number of the fuzzy integrators will depend of the number of sub modules or granules that the MNN has at a particular moment. The method was applied to the case of human recognition to illustrate its applicability with good results.
more …
By
Melin, Patricia; Leal, Ileana; Ochoa, Valente; Valenzuela, Luis; Torres, Gabriela; Clemente, Daniel
Show all (6)
We describe in this paper the application of several neural network architectures to the problem of simulating and predicting the dynamic behavior of complex economic time series. We use several neural network models and training algorithms to compare the results and decide at the end, which one is best for this application. We also compare the simulation results with the traditional approach of using a statistical model. In this case, we use real time series of prices of consumer goods to test our models. Real prices of tomato and green onion in the U.S. show complex fluctuations in time and are very complicated to predict with traditional statistical approaches.
more …
By
Trujillo, Leonardo; Martínez, Yuliana; Melin, Patricia
1 Citations
A fundamental task that must be addressed before classifying a set of data, is that of choosing the proper classification method. In other words, a researcher must infer which classifier will achieve the best performance on the classification problem in order to make a reasoned choice. This task is not trivial, and it is mostly resolved based on personal experience and individual preferences. This paper presents a methodological approach to produce estimators of classifier performance, based on descriptive measures of the problem data. The proposal is to use Genetic Programming (GP) to evolve mathematical operators that take as input descriptors of the problem data, and output the expected error that a particular classifier might achieve if it is used to classify the data. Experimental tests show that GP can produce accurate estimators of classifier performance, by evaluating our approach on a large set of 500 twoclass problems of multimodal data, using a neural network for classification. The results suggest that the GP approach could provide a tool that helps researchers make a reasoned decision regarding the applicability of a classifier to a particular problem.
more …
By
Valdez, Fevrier; Melin, Patricia; Castillo, Oscar
5 Citations
We describe in this paper an approach for mathematical function optimization using fuzzy logic for parameter tuning combining Particle Swarm Optimization (PSO) and Genetic Algorithms (GAs). The proposed method combines the advantages of PSO and GA to give us an improved FPSO+FGA hybrid method. Fuzzy logic is helpful to find the optimal parameters in PSO and GA in the best way possible. Also, with the tuning of parameters based on fuzzy logic it is possible to balance the exploration and exploitation of the proposed method. The hybrid method is called FPSO+FGA and was tested with a set of benchmark mathematical functions.
more …
By
Valdez, Fevrier; Melin, Patricia; Castillo, Oscar
1 Citations
We describe in this paper a new hybrid approach for mathematical function optimization combining Particle Swarm Optimization (PSO) and Genetic Algorithms (GAs) using Fuzzy Logic to integrate the results. The new evolutionary method combines the advantages of PSO and GA to give us an improved PSO+GA hybrid method. Fuzzy Logic is used to combine the results of the PSO and GA in the best way possible. The new hybrid PSO+GA approach is compared with the PSO and GA methods with a set of benchmark mathematical functions. The new hybrid PSO +GA method is shown to be superior than the individual evolutionary methods.
more …
By
Melin, Patricia; Gonzalez, Claudia; Bravo, Diana; Gonzalez, Felma; Martinez, Gabriela
Show all (5)
2 Citations
Abstract
We describe in this paper a new approach for pattern recognition using modular neural networks with a fuzzy logic method for response integration. We proposed a new architecture for modular neural networks for achieving pattern recognition in the particular case of human faces and fingerprints. Also, the method for achieving response integration is based on the fuzzy Sugeno integral with some modifications. Response integration is required to combine the outputs of all the modules in the modular network. We have applied the new approach for fingerprint and face recognition with a real database from students of our institution.
more …
By
Valdez, Fevrier; Melin, Patricia; Castillo, Oscar
7 Citations
We describe in this paper a Parallel Particle Swarm Optimization (PPSO) method with dynamic parameter adaptation to optimize complex mathematical functions. Fuzzy Logic is used to adapt the parameters of the PSO in the best way possible. The PPSO is shown to be superior to the individual evolutionary methods on the set of benchmark functions.
more …
By
Lopez, Miguel; Melin, Patricia; Castillo, Oscar
8 Citations
We describe in this paper a new method for response integration in ensemble neural networks with Type1 Fuzzy Logic and Type2 Fuzzy Logic using Genetic Algorithms (GA’s) for optimization. In this paper we consider pattern recognition with ensemble neural networks for the case of fingerprints. An ensemble neural network of three modules is used. Each module is a local expert on person recognition based on their biometric measure (Pattern recognition for fingerprints). The Response Integration method of the ensemble neural networks has the goal of combining the responses of the modules to improve the recognition rate of the individual modules. Using GA’s to optimize the Membership Functions of The Type1 Fuzzy System and Type2 Fuzzy System we can improve the results of the fuzzy systems. We show in this paper the results of a type2 approach for response integration that improves performance over the type1 logic approaches.
more …
By
Montiel, Oscar; Castillo, Oscar; Melin, Patricia; Sepúlveda, Roberto
Show all (4)
Abstract
There exists no standard method for obtaining a nonlinear inputoutput model using external dynamic approach. In this work, we are using an evolutionary optimization method for estimating the parameters of an NFIR model using the Wiener model structure. Specifically we are using a Breeder Genetic Algorithm (BGA) with fuzzy recombination for performing the optimization work. We selected the BGA since it uses real parameters (it does not require any string coding), which can be manipulated directly by the recombination and mutation operators. For training the system we used amplitude modulated pseudo random binary signal (APRBS). The adaptive system was tested using sinusoidal signals.
more …
By
Melin, Patricia; Martinez, Gabriela; Gonzalez, Claudia; Bravo, Diana; Gonzalez, Felma
Show all (5)
We describe in this paper a new approach for pattern recognition using modular neural networks with a fuzzy logic method for response integration. We proposed a new architecture for modular neural networks for achieving pattern recognition in the particular case of human fingerprints. Also, the method for achieving response integration is based on the fuzzy Sugeno integral. Response integration is required to combine the outputs of all the modules in the modular network. We have applied the new approach for fingerprint recognition with a real database of fingerprints from students of our institution.
more …
By
Hidalgo, Denisse; Melin, Patricia; Licea, Guillerrno; Castillo, Oscar
Show all (4)
6 Citations
We describe in this paper a new evolutionary method for the optimization of a modular neural network for multimodal biometry The proposed evolutionary method produces the best architecture of the modular neural network (number of modules, layers and neurons) and fuzzy inference systems (memberships functions and rules) as fuzzy integration methods. The integration of responses in the modular neural network is performed by using type1 and type2 fuzzy inference systems.
more …
By
Sepulveda, Roberto; Melin, Patricia
Abstract
applications. The use of new methods for handling incomplete information is of fundamental importance in engineering applications. This paper deals with the design of controllers using type2 fuzzy logic for minimizing the effects of uncertainty produced by the instrumentation elements. We simulated type1 and type2 fuzzy logic controllers to perform a comparative analysis of the systems’ response, in the presence of uncertainty. Uncertainty is an inherent part in controllers used for realworld
more …
By
Serrano, Rogelio; Tapia, Juan; Montiel, Oscar; Sepúlveda, Roberto; Melin, Patricia
Show all (5)
4 Citations
Multicore computers give the opportunity to solve highperformance applications more efficiently by using parallel computing. In this way, it is possible to achieve the same results in less time compared to the nonparallel version. Since computers continue to grow on the number of cores, we need to make our parallel applications scalable. This paper shows how a Genetic Algorithm (GA) in a nonparallel version takes long time to solve an optimization problem; in comparison, using multicore parallel computing the processing time can be reduced significantly as the number of cores grows. The tests were made on a quadcore computer; a comparison of the speeding up in relation to the number of cores is shown.
more …
By
Sotirov, Sotir; Sotirova, Evdokia; Melin, Patricia; Castilo, Oscar; Atanassov, Krassimir
Show all (5)
3 Citations
Modular neural networks (MNN) are a tool that can be used for object recognition and identification. Usually the inputs of the MNN can be fed with independent data. However, there are certain limits when we may use MNN, and the number of the neurons is one of the major parameters during the implementation of the MNN. On the other hand, the greater number of neurons slows down the learning process. In the paper, we propose a method for removing the number of the inputs and, hence, the neurons, without removing the error between the target value and the real value obtained on the output of the MNN’s exit. The method uses the recently proposed approach of InterCriteria Analysis, based on index matrices and intuitionistic fuzzy sets, which aims to detect possible correlations between pairs of criteria. The coefficients of the positive and negative consonance can be combined for obtaining the best results and smaller number of the weight coefficients of the neural network.
more …
By
Cervantes, Leticia; Castillo, Oscar; Melin, Patricia; Valdez, Fevrier
Show all (4)
4 Citations
In this paper, simulation results with type1 fuzzy systems and a type2 fuzzy granular approach for intelligent control of nonlinear dynamical plants are presented. First, the proposed method for intelligent control using a type2 fuzzy granular approach is described. Then, the proposed method is illustrated with the benchmark case of three tank water level control. Finally, a comparison between a type1 fuzzy system and the type2 fuzzy granular system for water control is presented.
more …
By
Valdez, Fevrier; Melin, Patricia; Castillo, Oscar
This paper describes a hybrid approach for optimization combining Particle Swarm Optimization (PSO) and Genetic Algorithms (GAs) using Fuzzy Logic to integrate the results, the proposed method is called FPSO+FGA. The new hybrid FPSO+FGA approach is compared with the Simulated Annealing (SA), PSO, GA, Pattern Search (PS) methods with a set of benchmark mathematical functions.
more …
By
MartinezSoto, Ricardo; Castillo, Oscar; Aguilar, Luis T.; Melin, Patricia
Show all (4)
11 Citations
In this paper we apply to Bioinspired and evolutionary optimization methods to design fuzzy logic controllers (FLC) to minimize the steady state error of linear systems. We test the optimal FLC obtained by the genetic algorithms and the PSO applied on linear systems using benchmark plants. The bioinspired and the evolutionary methods are used to find the parameters of the membership functions of the FLC to obtain the optimal controller. Simulation results are obtained with Simulink showing the feasibility of the proposed approach.
more …
By
Lopez, Miguel; Melin, Patricia; Castillo, Oscar
We describe in this paper a new method for response integration in ensemble neural networks with Type1 Fuzzy Logic and Type2 Fuzzy Logic using Genetic Algorithms (GA’s) for optimization. In this paper we consider pattern recognition with ensemble neural networks for the case of fingerprints to the test proposed method of response integration. An ensemble neural network of three modules is used. Each module is a local expert on person recognition based on their biometric measure (Pattern recognition for fingerprints). The Response Integration method of the ensemble neural networks has the goal of combining the responses of the modules to improve the recognition rate of the individual modules. First we use GA’s to optimize the fuzzy rules of The Type1 Fuzzy System and Type2 Fuzzy System to test the proposed method of response integration and after using GA’s to optimize the membership function of The Type1 Fuzzy Logic and Type2 Fuzzy logic to test the proposed method of response integration and finally show the comparison of the results between these methods. We show in this paper a comparative study of fuzzy methods for response integration and the optimization of the results of a type2 approach for response integration that improves performance over the type1 logic approaches.
more …
By
Cervantes, Leticia; Castillo, Oscar; Melin, Patricia
14 Citations
In this paper we present simulation results that we have at this moment with a new approach for intelligent control of nonlinear dynamical plants. First we present the proposed approach for intelligent control using a hierarchical modular architecture with type2 fuzzy logic used for combining the outputs of the modules. Then, the approach is illustrated with two cases: aircraft control and shower control and in each problem we explain its behavior. Simulation results of the two case show that proposed approach has potential in solving complex control problems.
more …
By
Hidalgo, Denisse; Melin, Patricia; Castillo, Oscar
1 Citations
In this paper we describe a method for the optimization of type2 fuzzy systems based on the level of uncertainty considering three different cases to reduce the complexity problem of searching the solution space. The proposed method produces the best fuzzy inference systems for particular applications based on a genetic algorithm. We apply a Genetic Algorithm to find the optimal type2 fuzzy system dividing the search space in three subspaces. We show the comparative results obtained for the benchmark problems.
more …
By
Mendoza, Olivia; Melin, Patricia; Castillo, Oscar; Licea, Guillermo
Show all (4)
19 Citations
The combination of Soft Computing techniques allows the improvement of intelligent systems with different hybrid approaches. In this work we consider two parts of a Modular Neural Network for image recognition, where a Type2 Fuzzy Inference System (FIS 2) makes a great difference. The first FIS 2 is used for feature extraction in training data, and the second one to find the ideal parameters for the integration method of the modular neural network. Once again Fuzzy Logic is shown to be a tool that can help improve the results of a neural system, when facilitating the representation of the human perception.
more …
By
Montiel, Oscar; Castillo, Oscar; Melin, Patricia; Sepúlveda, Roberto
Show all (4)
Abstract
There exists no standard method for obtaining a nonlinear inputoutput model using external dynamic approach. In this work, we are using an evolutionary optimization method for estimating the parameters of an NFIR model using the Wiener model structure. Specifically we are using a Breeder Genetic Algorithm (BGA) with fuzzy recombination for performing the optimization work. We selected the BGA since it uses real parameters (it does not require any string coding), which can be manipulated directly by the recombination and mutation operators. For training the system we used amplitude modulated pseudo random binary signal (APRBS). The adaptive system was tested using sinusoidal signals.
more …
By
Castillo, Oscar; Melin, Patricia; Valdez, Fevrier
A review of the optimization methods used in the design of type2 fuzzy systems, which are relatively novel models of imprecision, is presented in this paper. The main aim of the work is to study the basic reasons for optimizing type2 fuzzy systems for solving problems different areas of application. Recently, natureinspired methods have emerged as powerful optimization algorithms for solving complex problems. In the case of designing type2 fuzzy systems for particular applications, the use of natureinspired optimization methods have helped in the complex task of finding the appropriate parameter values and structure of the fuzzy systems. In this paper, we consider the application of genetic algorithms, particle swarm optimization and ant colony optimization as three different paradigms that help in the design of optimal type2 fuzzy systems. A comparison of the different optimization methods for the case of designing type2 fuzzy systems is also offered.
more …
By
Castro, Juan R.; Castillo, Oscar; Melin, Patricia; RodríguezDíaz, Antonio
Show all (4)
26 Citations
This paper presents the development and design of a graphical user interface and a command line programming Toolbox for construction, edition and simulation of Interval Type2 Fuzzy Inference Systems. The Interval Type2 Fuzzy Logic System (IT2FLS) Toolbox, is an environment for interval type2 fuzzy logic inference system development. Tools that cover the different phases of the fuzzy system design process, from the initial description phase, to the final implementation phase, constitute the Toolbox. The Toolbox’s best qualities are the capacity to develop complex systems and the flexibility that allows the user to extend the availability of functions for working with the use of type2 fuzzy operators, linguistic variables, interval type2 membership functions, defuzzification methods and the evaluation of Interval Type2 Fuzzy Inference Systems.
more …
By
Serrano, Magdalena; Ayala, Erika; Melin, Patricia
This paper presents an intelligent system for person identification with biometric measures such as signature, fingerprint and face. We describe the neural network architectures used to achieve person identification based on the biometrics measures. Simulation results show that the proposed method provides good recognition.
more …
By
Soto, Jesus; Melin, Patricia; Castillo, Oscar
1 Citations
This paper describes the Particle Swarm Optimization of the Fuzzy integrators in Ensembles of ANFIS (adaptive neurofuzzy inferences systems) models for the prediction time series. A chaotic system is considered in this work, which is the MackeyGlass time series, that is generated from a model is in the form of differential equations. This benchmark time series is used to test of performance of the proposed optimization of the fuzzy integrators in ensemble architecture. We used interval type2 and type1 fuzzy systems to integrate the output (forecast) of each Ensemble of ANFIS models. Particle Swarm Optimization (PSO) was used for the optimization of membership function parameters of each fuzzy integrator. In the experiments we optimized Gaussian, Generalized Bell and Triangular membership functions parameters for each of the fuzzy integrators. Simulation results show the effectiveness of the proposed approach. Therefore, a comparison was made against another recent work to validate the performance of the proposed model.
more …
By
Melin, Patricia; Mancilla, Alejandra; Lopez, Miguel; Solano, Daniel; Soto, Miguel; Castillo, Oscar
Show all (6)
2 Citations
We describe in this paper the evolution of modular neural networks using hierarchical genetic algorithms for pattern recognition. Modular Neural Networks (MNN) have shown significant learning improvement over single Neural Networks (NN). For this reason, the use of MNN for pattern recognition is well justified. However, network topology design of MNN is at least an order of magnitude more difficult than for classical NNs. We describe in this paper the use of a Hierarchical Genetic Algorithm (HGA) for optimizing the topology of each of the neural network modules of the MNN. The HGA is clearly needed due to the fact that topology optimization requires that we are able to manage both the layer and node information for each of the MNN modules. Simulation results prove the feasibility and advantages of the proposed approach.
more …
By
González, Beatriz; Valdez, Fevrier; Melin, Patricia
In this paper we are presenting a modification of the Gravitational Search Algorithm (GSA) using type2 fuzzy logic to dynamically change the alpha parameter and provide a different gravitation and acceleration values to each agent in order to improve its performance. We test this approach with benchmark mathematical functions. Simulation results show the advantages of the proposed approach.
more …
By
Beltrán, Mónica; Melin, Patricia; Trujillo, Leonardo
2 Citations
This chapter describes a modular neural network (MNN) for the problem of signature recognition. Currently, biometric identification has gained a great deal of research interest within the pattern recognition community. For instance, many attempts have been made in order to automate the process of identifying a person’s handwritten signature, however this problem has proven to be a very difficult task. In this work, we propose a MNN that has three separate modules, each using different image features as input, these are: edges, wavelet coefficients, and the Hough transform matrix. Then, the outputs from each of these modules are combined using a Sugeno fuzzy integral. The experimental results obtained using a database of 30 individual’s shows that the modular architecture can achieve a very high 98% recognition accuracy with a test set of 150 images. Therefore, we conclude that the proposed architecture provides a suitable platform to build a signature recognition system.
more …
By
Pulido, Martha; Mancilla, Alejandra; Melin, Patricia
2 Citations
In this paper we describe the application of the architecture for an ensemble neural network for Complex Time Series Prediction. The time series we are considering is the MackeyGlass, and we show the results of some simulations with the ensemble neural network, and its integration with the methods of average, weighted average and Fuzzy Integration. Simulation results show very good prediction of the ensemble neural network with fuzzy integration.
more …
By
Montiel, Oscar; Castillo, Oscar; Melin, Patricia; Sepulveda, Roberto
Show all (4)
1 Citations
In this work we are optimizing an adaptive finite impulse response filter applied to the problem of system identification. We are proposing a breeder genetic algorithm (BGA) for performing the parametric search in highly multimoldal landscapes since in this kind of filters there exits epistiasis. The results of BGA were compared to the traditional genetic algorithm, and we found that the BGA was clearly superior (in accuracy) in most of the cases. We used the statistical least mean squared for validating the results. We suggest to hybridized both methods for real world applications.
more …
By
Valdez, Fevrier; Melin, Patricia; Castillo, Oscar
2 Citations
We describe in this paper a new hybrid approach for mathematical function optimization combining Particle Swarm Optimization (PSO) and Genetic Algorithms (GAs) using Fuzzy Logic to integrate the results. The new evolutionary method combines the advantages of PSO and GA to give us an improved PSO+GA hybrid method. Fuzzy Logic is used to combine the results of the PSO and GA in the best way possible. The new hybrid PSO+GA approach is compared with the PSO and GA methods with a set of benchmark mathematical functions. The new hybrid PSO+GA method is shown to be superior than the individual evolutionary methods.
more …
By
Méndez, Eduardo; Castillo, Oscar; Soria, José; Melin, Patricia; Sadollah, Ali
Show all (5)
1 Citations
This paper describes the enhancement of the Water Cycle Algorithm (WCA) using a fuzzy inference system to adapt its parameters dynamically. The original WCA is compared regarding performance with the proposed method called Water Cycle Algorithm with Dynamic Parameter Adaptation (WCADPA). Simulation results on a set of wellknown test functions show that the WCA can be improved with a fuzzy dynamic adaptation of the parameters.
more …
By
Montiel, Oscar; Castillo, Oscar; Melin, Patricia; Sepulveda, Roberto
Show all (4)
1 Citations
Summary
In this chapter, we are proposing an approach for integrating evolutionary computation applied to the problem of system identification in the wellknown statistical signal processing theory. Here, some mathematical expressions are developed in order to justify the learning rule in the adaptive process when a Breeder Genetic Algorithm is used as the optimization technique. In this work, we are including an analysis of errors, energy measures, and stability.
more …
By
Valdez, Fevrier; Melin, Patricia; Licea, Guillermo
2 Citations
We describe in this paper a new hybrid approach for optimization combining Particle Swarm Optimization (PSO) and Genetic Algorithms (GAs) using Fuzzy Logic to integrate the results. The new evolutionary method combines the advantages of PSO and GA to give us an improved FPSO+FGA hybrid method. Fuzzy Logic is used to combine the results of the PSO and GA in the best way possible. Also fuzzy logic is used to adjust parameters in the FPSO and FGA. The new hybrid FPSO+FGA approach is compared with the PSO and GA methods for the optimization of modular neural networks. The new hybrid FPSO+FGA method is shown to be superior with respect to both the individual evolutionary methods.
more …
By
Muñoz, Ricardo; Castillo, Oscar; Melin, Patricia
5 Citations
In this paper we describe a Modular Neural Network (MNN) with fuzzy integration for face, fingerprint and voice recognition. The proposed MNN architecture defined in this paper consists of three modules; face, fingerprint and voice. Each of the mentioned modules is divided again into three sub modules. The same information is used as input to train the sub modules. Once we have trained and tested the MNN modules, we proceed to integrate these modules with a fuzzy integrator. In this paper we demonstrate that using MNNs for face, fingerprint and voice recognition integrated with a fuzzy integrator is a good option to solve pattern recognition problems.
more …
By
Castillo, Oscar; Melin, Patricia
Article Outline
Glossary
Definition of the Subject
Introduction
Genetic Algorithm for Optimization
Evolution of Fuzzy Systems
Application to Anesthesia Control
Application to the Control of the Bar and Ball System
Hierarchical Genetic Algorithms for Neural Networks
Experimental Results for Time Series Prediction
Conclusions
Future Directions
Bibliography
more …
By
Peraza, Cinthia; Valdez, Fevrier; Castillo, Oscar; Melin, Patricia
Show all (4)
A fuzzy harmony search algorithm (FHS) is presented in this paper. This method uses a fuzzy system for dynamic adaptation of the harmony memory accepting (HMR) parameter along the iterations, and in this way achieving control of the intensification and diversification of the search space. This method was previously applied to classic benchmark mathematical functions with different number of dimensions. However, in this case we decided to apply the proposed FHS to benchmark mathematical problems provided by the CEC 2015 competition, which are unimodal, multimodal, hybrid and composite functions to check the efficiency for the proposed method. A comparison is presented to verify the results obtained with respect to the original harmony search algorithm and fuzzy harmony search algorithm.
more …
By
Leal, Ileana; Melin, Patricia
Abstract
In this paper we describe the concepts of Time Series, Neural Networks, Modular Neural Networks, and Parallelism. Modular Neural Networks and Parallel Processing for Time Series Forecasting of Tomato Prices in Mexico are described in this paper. A particular modular neural network architecture implemented in parallel was used. Simulation results with the modular neural network approach for this application are very good.
more …
By
Villegas, José M.; Mancilla, Alejandra; Melin, Patricia
This paper deals with two optimization problems as the architecture (modules, layers and neurons) and the best training of an artificial neural network (ANN). For that matter is used a Hierarchical Genetic Algorithm, which theorically has the capacity to bring the optimal architecture and the training result of the ANN, for a particular task; in this case the recognition of an individual is via voice and face.
more …
By
Melin, Patricia; Pulido, Martha; Castillo, Oscar
1 Citations
This paper describes the design of ensemble neural networks using Particle Swarm Optimization (PSO) for time series prediction with Type1 and Type2 Fuzzy Integration. The time series that is being considered in this work is the MackeyGlass benchmark time series. Simulation results show that the ensemble approach produces good prediction of the MackeyGlass time series.
more …
By
Amezcua, Jonathan; Melin, Patricia
In this paper, the optimization of LVQ neural networks with modular approach is presented for classification of arrhythmias, using particle swarm optimization. This work focuses only in the optimization of the number of modules and the number of cluster centers. Other parameters, such as the learning rate or number of epochs are static values and are not optimized. Here, the MITBIH arrhythmia database with 15 classes was used. Results show that using 5 modules architecture could be a good approach for classification of arrhythmias.
more …
By
Lagunes, Marylu L.; Castillo, Oscar; Valdez, Fevrier; Soria, Jose; Melin, Patricia
Show all (5)
This paper describes the comparison of dynamic adjustment parameters in the firefly algorithm using type1 and type2 fuzzy logic for the optimization of a fuzzy controller. The adjustment is performed to improve the behavior of the method. Fuzzy systems use fuzzy sets by defining membership functions, which indicate how much an element belongs to the fuzzy set. Type2 fuzzy logic assigns degrees of belonging that are fuzzy and this can be viewed as an extension of type1 fuzzy logic. The Firefly algorithm has 3 main parameters Beta, Gamma and Alpha with a range of 0 to 1 each, which need to the dynamically adjusted to improve the performance of the algorithm.
more …
By
SalazarTejeda, Pedro Antonio; Melin, Patricia; Castillo, Oscar
13 Citations
In this paper, we describe an application of biometric recognition that is structured basically with three inputs: the hand geometry, voice and image. The hand geometry is given by an image of “the palm” of the hand with a 480x640 size which is preprocessed with a feature extraction that uses computer vision techniques and with certain features we recognize the individual. After that we preprocessed the image and get some variables as the fingers, palm, wrist, also a segment of the palm; they appear to be from a with a fuzzy system that will tell us how much they seemed to a certain person, comparing each variable given by the preprocessing of the image according to the data base that its already stored (all the images of the individuals, voice, etc.).
more …
By
Guzmán, Juan Carlos; Melin, Patricia; PradoArechiga, German
One of the most dangerous diseases for humans is the Arterial Hypertension, which this kind of disease that often leads to fatal outcomes, such as heart attack, stroke and renal failure. The hypertension seriously threats the health of the people worldwide. One of the dangerous aspects of the hypertension is that you may not know that you have it. In fact, nearly onethird of people who have high blood pressure don’t know it. The only way to know if the blood pressure is high is through the regular checkups. The evaluation of a patient with Hypertension should (1) confirm the diagnosis of hypertension, (2) detect causes of secondary hypertension y (3) assess cardio vascular risk and organ damage. Therefore, is very important a correct measurement of the blood pressure (BP). Traditionally, office BP measurement has been performed using a sphygmomanometer and stethoscope. Recently, automated office and home BP measurements has been proposed as an alternative to traditional measurement. It has several advantages over manual BP, especially in routine clinical practice. Therefore, we have developed a Fuzzy System for the diagnosis of the Hypertension. Firstly, the input parameters include Systolic Blood Pressure and Diastolic Blood Pressure. Secondly, we have as an output parameter: Blood Pressure Levels (BPL). The input linguistic value includes Low, Low Normal, Normal, High Normal, High, Very High, Too High and Isolated Systolic Hypertension. Finally, we have 14 fuzzy rules to determine the diagnosis output.
more …
By
Hidalgo, Denisse; Melin, Patricia; Licea, Guillermo
2 Citations
In this paper we describe a new evolutionary method to perform the optimization of a modular neural network applied to the case of multimodal biometry. Integration of responses in the modular neural network is performed using type1 and type2 fuzzy inference systems.
The proposed evolutionary method produces the best architecture of the modular neural network.
more …
By
Hidalgo, Denisse; Castillo, Oscar; Melin, Patricia
15 Citations
We describe in this paper a comparative study of Fuzzy Inference Systems as methods of integration in modular neural networks (MNN’s) for multimodal biometry. These methods of integration are based on type1 and type2 fuzzy logic. Also, the fuzzy systems are optimized with simple genetic algorithms. First, we considered the use of type1 fuzzy logic and later the approach with type2 fuzzy logic. The fuzzy systems were developed using genetic algorithms to handle fuzzy inference systems with different membership functions, like the triangular, trapezoidal and Gaussian; since these algorithms can generate the fuzzy systems automatically. Then the response integration of the modular neural network was tested with the optimized fuzzy integration systems. The comparative study of type1 and type2 fuzzy inference systems was made to observe the behavior of the two different integration methods f modular neural networks for multimodal biometry.
more …
By
Castro, Juan R.; Castillo, Oscar; Melin, Patricia; RodríguezDíaz, Antonio
Show all (4)
5 Citations
In this work, a class of Interval Type2 Fuzzy Neural Networks (IT2FNN) is proposed, which is functionally equivalent to interval type2 fuzzy inference systems. The computational process envisioned for fuzzy neural systems is as follows: it starts with the development of an ”Interval Type2 Fuzzy Neuron”, which is based on biological neural morphologies, followed by the learning mechanisms. We describe how to decompose the parameter set such that the hybrid learning rule of adaptive networks can be applied to the IT2FNN architecture for the TakagiSugenoKang reasoning.
more …
By
Mendivil, Salvador González; Castillo, Oscar; Melin, Patricia
2 Citations
This paper considers the application of parallel genetic algorithms to the optimization of modular neural network architectures for time series prediction. We have a cluster configuration of 16 computers and the application is executed using the Matlab Distributed Computing Engine included in MATLAB r2006b. The Linux Fedora Core VI Operating System was installed and configured for the cluster execution due to its high performance, scalability and because it presents innumerable benefits that facilitate the implementation of distributed computing applications. The first part of this paper presents the theoretical framework with basic concepts like times series, artificial neural networks, genetic algorithms, and parallel genetic algorithms. The second part of this paper presents the procedure for configuring the cluster of computers, requirements, experiences and main problems that were encountered. Also, the development of the project is presented explaining as it was initially proposed and the adjustments that were required. The third part of this paper presents the obtained results for the time series prediction using tables, graphics and describing each one of them. Finally the conclusions and future works are presented.
more …
