Skip to main content

Über dieses Buch

This book presents recent advances on hybrid intelligent systems using soft computing techniques for intelligent control and robotics, pattern recognition, time series prediction and optimization of complex problems. Soft Computing (SC) consists of several intelligent computing paradigms, including fuzzy logic, neural networks, and bio-inspired optimization algorithms, which can be used to produce powerful hybrid intelligent systems. The book is organized in five main parts, which contain groups of papers around a similar subject. The first part consists of papers with the main theme of hybrid intelligent systems for control and robotics, which are basically state of the art papers that propose new models and concepts, which can be the basis for achieving intelligent control and mobile robotics. The second part contains papers with the main theme of hybrid intelligent systems for pattern recognition and time series prediction, which are basically papers using nature-inspired techniques, like evolutionary algorithms, fuzzy logic and neural networks, for achieving efficient pattern recognition or time series prediction. The third part contains papers with the theme of bio-inspired and genetic optimization methods, which basically consider the proposal of new methods and applications of bio-inspired optimization to solve complex optimization of real problems. The fourth part contains papers that deal with the application of intelligent optimization techniques in real world problems in scheduling, planning and manufacturing. The fifth part contains papers with the theme of evolutionary methods and intelligent computing, which are papers considering soft computing methods for applications related to diverse areas, such as natural language processing, recommending systems and optimization.



Hybrid Intelligent Systems for Control and Robotics


Optimization of a Fuzzy Tracking Controller for an Autonomous Mobile Robot under Perturbed Torques by Means of a Chemical Optimization Paradigm

This paper addresses the tracking problem for the dynamic model of a unicycle mobile robot. A novel optimization method inspired on the chemical reactions is applied to solve this motion problem by integrating a kinematic and a torque controller based on fuzzy logic theory. Computer simulations are presented confirming that this optimization paradigm is able to outperform other optimization techniques applied to this particular robot application.

Leslie Astudillo, Patricia Melin, Oscar Castillo

Evolutionary Optimization of the Fuzzy Integrator in a Navigation System for a Mobile Robot

This paper describes the optimization of an Integrator control block within the proposed navigation control system for a mobile robot. The control blocks that the integrator will combine are two Fuzzy Inference Systems (FIS) in charge of tracking and reaction respectively. The integrator block is call Weighted Fussy Inference System (WFIS), and assigns weights to the responses on each behavior block, to combine them into a single response.

Abraham Meléndez, Oscar Castillo

Particle Swarm Optimization for Average Approximation of Interval Type-2 Fuzzy Inference Systems Design in FPGAs for Real Applications

This paper proposes the particle swarm optimization (PSO) of type-2 membership functions for average approximation of an interval type-2 fuzzy logic system (AT2-FIS), the AT2-FIS is synthesized in VHDL code for FPGAs. The PSO method considers three objective functions, overshoot, undershoot and steady state error because the real application is to control the speed of a DC motor. Several experiments were performed to optimize the AT2-FIS in FPGA. Experiments were conducted by changing the number of bits for encoding the AT2-FLC in VHDL.

Yazmin Maldonado, Oscar Castillo, Patricia Melin

Genetic Optimization of Membership Functions in Modular Fuzzy Controllers for Complex Problems

In this paper a method to design modular fuzzy controllers using genetic optimization is presented. The method is tested with a problem that requires 5 individual controllers. Simulation results with a genetic algorithm for optimizing membership functions of the 5 controllers are presented. Simulation results show that the proposed modular control approach offers advantages over existing control methods.

Leticia Cervantes, Oscar Castillo

Designing Systematic Stable Fuzzy Logic Controllers by Fuzzy Lyapunov Synthesis

Fuzzy logic handles information imprecision using intermediate expressions to define assessments. Fuzzy Systems are intelligent models whose main application has been in Control Engineering applications. Stability is one of the most important issues of control systems. This determines the system to respond in an acceptable way. This work is based on the fuzzy Lyapunov synthesis in the design of fuzzy controllers, to verify the system’s stability. The stability will be studied on Mamdani and Sugeno fuzzy systems .The case study presented is a system of a cylindrical tank of water, where we aim to maintain a certain level of water, which is regulated through the controls applied to the water outlet valve of the tank. The method is also tested using an inverted pendulum, which is an unstable system, which can fall at any time unless an appropriate force is applied control.

María Concepción Ibarra, Oscar Castillo, Jose Soria

Design of Fuzzy Control Systems with Different PSO Variants

This paper describes the metaheuristic of Optimization by Swarm of Particles (PSO-Particle Swarm Optimization) and its variants (Clamping speed, inertia and constriction coefficient) as an optimization strategy to design the membership functions of Benchmark Control Cases (Tank water and Inverted Pendulum) Each of the variants have their own advantages within the algorithm because they allow the exploration and exploitation in different ways and this allows us to find the optimum.

Resffa Fierro, Oscar Castillo

Methodology to Design Fuzzy Logic Controller for Soft-Core Embedded into FPGA

In this paper, the methodology for the design of fuzzy controllers for softcore processors, such as the Xilinx Microblaze embedded in the VIRTEX5 FPGA, is proposed to regulate the angular position of the axes of an experimental platform. The platform uses servomotors to control the rotational movements of the X-Y-Z axes, this with respect to the earth horizon. The angular position is feedback using three inclinometers sensors based on MEMS technology with SPI interface. The desired position is regulated using three independent fuzzy PD+I controllers, which use the error and change of error as input signals. The proposed methodology consists in the design and evaluation of the fuzzy controllers using the Fuzzy Logic Toolbox of Matlab.

Roberto Sepúlveda, Oscar Montiel-Ross, Jorge Quiñones

Optimization of Membership Functions for Type-1 and Type 2 Fuzzy Controllers of an Autonomous Mobile Robot Using PSO

This paper describes the application of the optimization algorithm based on particle swarms known by its acronym as PSO, used to adjust the parameters of membership functions of a fuzzy logic controller (FLC) to find the optimal intelligent control for a wheeled autonomous mobile robot. Results of several simulations show that the PSO is able to optimize the type-1 and type 2 FLCs for this specific application.

San Juana Aguas-Marmolejo, Oscar Castillo

Hybrid Intelligent Systems for Pattern Recognition and Time Series Prediction


Multi-Objective Hierarchical Genetic Algorithm for Modular Neural Network Optimization Using a Granular Approach

In this paper we propose a multi-objective hierarchical genetic algorithm (MOHGA) for modular neural network optimization. A granular approach is used due to the fact that the dataset is divided into granules or sub modules. The main objective of this method is to know the optimal number of sub modules or granules, but also allow the optimization of the number of hidden layers, number of neurons per hidden layer, error goal and learning algorithms per module. The proposed MOHGA is based on the Micro genetic algorithm and was tested for a pattern recognition application. Simulation results show that the proposed modular neural network approach offers advantages over existing neural network models.

Daniela Sánchez, Patricia Melin

An Analysis on the Intrinsic Implementation of the Principle of Justifiable Granularity in Clustering Algorithms

The initial process for the granulation of information is the clustering of data, once the relationships between this data have been found these become clusters, each cluster represents a coarse granule, whereas each data point represents a fine granule. All clustering algorithms find these relationships by different means, yet the notion of the principle of justifiable granularity is not considered by any of them, since it is a recent idea in the area of Granular Computing. This paper describes a first approach in the analysis of the relationship between the size of the clusters found and their intrinsic implementation of the principle of justifiable granularity. An analysis is done with two datasets, simplefit and iris, and two clustering algorithms, subtractive and granular gravitational.

Mauricio A. Sanchez, Oscar Castillo, Juan R. Castro

Mental Tasks Temporal Classification Using an Architecture Based on ANFIS and Recurrent Neural Networks

In this paper, an architecture based on adaptive neuro-fuzzy inference systems (ANFIS) assembled to recurrent neural networks, applied to the problem of mental tasks temporal classification, is proposed. The electroencephalographic signals (EEG) are pre-processed through band-pass filtering in order to separate the set of energy signals in alpha and beta bands. The energy in each band is represented by fuzzy sets obtained through an ANFIS system, and the temporal sequence corresponding to the combination to be detected, associated to the specific mental task, is entered into a recurrent neural networks. This experiment has been carried out in the context of brain-computer-interface (BCI) systems development. Experimentation using EEG signals corresponding to mental tasks exercises, obtained from a database available to the international community for research purposes, is reported. Two recurrent neural networks are used for comparison purposes: Elman network and a fully connected recurrent neural network (FCRNN) trained by RTRL-EKF (real time recurrent learning – extended Kalman filter). A classification rate of 88.12% in average was obtained through the FCRNN during the generalization stage.

Emmanuel Morales-Flores, Juan Manuel Ramírez-Cortés, Pilar Gómez-Gil, Vicente Alarcón-Aquino

Interval Type-2 Fuzzy System for Image Edge Detection Quality Evaluation Applied to Synthetic and Real Images

In this paper a new method to calculate a quality index for edge detection of an image is proposed. The new method can be applied to synthetic or real images, and consists on an interval type-2 fuzzy system (IT2FS). The inputs for the IT2FS correspond to a combination of parameters representing the most influential characteristics of an edge image according on literature and our previous experience. This new index can be calculated for any edge detected image, including the traditional and fuzzy methods.

Felicitas Perez-Ornelas, Olivia Mendoza, Patricia Melin, Juan R. Castro

Genetic Optimization of Type-2 Fuzzy Weight Adjustment for Backpropagation in Ensemble Neural Network

In this paper a genetic algorithm is used to optimize the three neural networks forming an ensemble. Genetic algorithms are also used to optimize the two type-2 fuzzy systems that work in the backpropagation learning method with type-2 fuzzy weight adjustment. The mathematical analysis of the proposed learning method architecture and the adaptation of type-2 fuzzy weights are presented. The proposed method is based on recent methods that handle weight adaptation and especially fuzzy weights. In this work an ensemble neural network of three neural networks and average integration to obtain the final result is presented. The proposed approach is applied to a case of time series prediction.

Fernando Gaxiola, Patricia Melin, Fevrier Valdez

A New Method for Type-2 Fuzzy Integration in Ensemble Neural Networks Based on Genetic Algorithms

This paper describes a proposed method for type-2 fuzzy integration that can be used in the fusion of responses for an ensemble neural network. We consider the case of the design of a type-2 fuzzy integrator for fusion of a neural network ensemble. The network structure of the ensemble may have a maximum of 5 modules. This integrator consists of 32 fuzzy rules, with 5 inputs depending on the number of modules of the neural network ensemble and one output. Each input and output linguistic variable of the fuzzy system uses Gaussian membership functions. The performance of type-2 fuzzy integrators is analyzed under different levels of uncertainty to find out the best design of the membership functions. In this case the proposed method is applied to time series prediction.

Martha Pulido, Patricia Melin

A Hand Geometry Biometric Identification System Utilizing Modular Neural Networks with Fuzzy Integration

The present work deals with the problem of identifying individuals from a database, and in so doing utilizing measurements taken from handpalm images. The techniques utilized for performing identifications are mainly those of artificial neural networks, which work upon the data through the use of two modular neural networks, one which is concerned solely with the handpalm image, another with the measurements taken thereof. Outputs from these two networks are integrated through a fuzzy inference system. Subsequent work will comprise improvement of the obtained results.

José Luis Sánchez, Patricia Melin

Development of an Automatic Method for Classification of Signatures in a Recognition System Based on Modular Neural Networks

This paper presents the development of an automatic method for classification of signatures according to a geometric shape recognition system based on a modular neural network (MNN). To carry out the method of automatic classification, we apply pre-processing to our database which consists of 30 individuals, first the image goes to a high pass filter, which allows the passage of signals depending on their frequency and finally we apply the Fourier transform, which is essentially a wave phenomenon which serves to measure the distribution of amplitudes of the frequency of our image (signatures), and presents to certain extent as they are, rise time, peak parking. Thus the signatures are automatically sorted to the module that corresponds in the Modular Neural Network, which contains three separate modules, each one uses different feature extraction methods: edge extraction, wavelet transform and Hough transform, where this results in the identification or recognition of signatures.

Verónica Carrera, Patricia Melin, Diana Bravo

Architecture of Modular Neural Network in Pattern Recognition

This paper introduces on architecture of a modular neural network (MNN) for pattern recognition, more recently, the addition of modular neural network techniques theory have been receiving significant attention. The design of a recognition system requires careful. The paper also aims to use the architecture of this Modular Neural Network for pattern recognition in order to optimize the architecture, and used an integrator that will get a good percentage of image identification and in the shortest time possible.

Manuel Leobardo Zavala-Arriaza, Fevrier Valdez, Patricia Melin

Bio-Inspired and Genetic Optimization Methods

Comparative Study of Particle Swarm Optimization Variants in Complex Mathematics Functions

Particle Swarm Optimization (PSO) is one of the evolutionary computation techniques based on the social behaviors of birds flocking or fish schooling, biologically inspired computational search and optimization method. Since first introduced by Kennedy and Eberhart [7] in 1995, several variants of the original PSO have been developed to improve speed of convergence, improve the quality of solutions found, avoid getting trapped in the local optima and so on. This paper is focused on performing a comparison of different PSO variants such as full model, only cognitive, only social, weight inertia, and constriction factor. We are using a set of 4 mathematical functions to validate our approach. These functions are widely used in this field of study.

Juan Carlos Vazquez, Fevrier Valdez, Patricia Melin

A Method to Solve the Traveling Salesman Problem Using Ant Colony Optimization Variants with Ant Set Partitioning

In this paper we propose an ant’s partition method for Ant Colony Optimization (ACO), a meta-heuristic that is inspired in ant’s behavior and how they collect their food. The proposed method equivalently divides the total number of ants in three different subsets and each one is evaluated separately by the corresponding variation of ACO (AS, EAS, MMAS) to solve different instances of The Traveling Salesman Problem (TSP). This method is based on the idea of “divide and conquer” to be applied in the division of the work, as the ants are evaluated in different ways in the same iteration. This method also includes a stagnation mechanism that stops at a certain variation if it’s not working properly after several iterations. This allows us to save time performing tests and have less overhead in comparison with the conventional method, which uses just one variation of ACO in all iterations.

Evelia Lizárraga, Oscar Castillo, José Soria

Particle Swarm Optimization with Dynamic Parameter Adaptation Using Fuzzy Logic for Benchmark Mathematical Functions

In this paper a new method for dynamic parameter adaptation in particle swarm optimization (PSO) is proposed. PSO is a metaheuristic inspired in social behaviors, which is very useful in optimization problems. In this paper we propose an improvement to the convergence and diversity of the swarm in PSO using fuzzy logic. Simulation results show that the proposed approach improves the performance of PSO.

Frumen Olivas, Oscar Castillo

Dynamic Fuzzy Logic Parameter Tuning for ACO and Its Application in TSP Problems

Ant Colony Optimization (ACO) is a population-based constructive metaheuristic that exploits a form of past performance memory inspired by the foraging behavior of real ants. The behavior of the ACO algorithm is highly dependent on the values defined for its parameters. Adaptation and parameter control are recurring themes in the field of bio-inspired algorithms. The present paper explores a new approach of diversity control in ACO. The central idea is to avoid or slow down full convergence through the dynamic variation of the alpha parameter. The performance of different variants of the ACO algorithm was observed to choose one as the basis to the proposed approach. A convergence fuzzy logic controller with the objective of maintaining diversity at some level to avoid premature convergence was created. Encouraging results on several travelling salesman problem (TSP) instances are presented with the proposed method.

Héctor Neyoy, Oscar Castillo, José Soria

Using Immunogenetic Algorithms for Solving Combinatorial Optimization Problems

We present a novel approach for reducing the computational time of Combinatorial Optimization Problems (COPs). The approach is inspired by the use of Artificial Vaccines, a concept that is classified as being part of a set of algorithms called Artificial Immune Systems (AIS). Artificial Vaccines are able to reduce the computational time and quality of the solution of any existing COP solving algorithm by reducing the size of the problem set. To demonstrate the usefulness of this proposal we provide comparative results obtained with a Genetic Algorithm (GA) solving the Traveling Salesman Problem (TSP) for three large instances of 423, 737 and 1583 cities.

Francisco Javier Díaz-Delgadillo, Oscar Montiel-Ross, Roberto Sepúlveda

Comparison of Metaheuristic Algorithms with a Methodology of Design for the Evaluation of Hard Constraints over the Course Timetabling Problem

The Course Timetabling problem is one of the most difficult and common problems inside an university. The main objective of this problem is to obtain a timetabling with the minimum student conflicts between assigned activities. A Methodology of design is a strategy applied before the execution of an algorithm for timetabling problem. This strategy has recently emerged, and aims to improve the obtained results as well as provide a context-independent layer to different versions of the timetabling problem. This methodology offers to an interested researcher the advantage of solving different set instances with a single algorithm which is a new paradigm in the timetabling problem state of art. In this paper the proposed methodology is tested with several metaheuristic algorithms over some well-know set instances such as Patat 2002 and 2007. The main objective in this work is to find which metaheuristic algorithm shows a better performance in terms of quality, used together with the Design Methodology. The algorithms chosen are from the area of evolutionary computation, Cellular algorithms and Swarm Intelligence. Finally our experiments use some non-parametric statistical test like Kruskal-Wallis test and wilcoxon signed rank test.

A. Soria-Alcaraz Jorge, Carpio Martín, Puga Héctor, Marco Aurelio Sotelo-Figueroa

Path Planning Using Clonal Selection Algorithm

We used the clonal selection algorithm (CSA) to generate optimal or nearly optimal solutions for solving combinatorial optimization problems, the algorithm was applied to generate quality G-Code sequences for a CNC Mill machine tool problem. We validated the software using the Traveler Salesman Problem (TSP) and demonstrated that the algorithm can optimize the travel path by reducing manufacturing time and costs. Different optimization experiments using the CSA of manufacturing codes of application CAD/CAM software are illustrated.

Christian Huizar, Oscar Montiel-Ross, Roberto Sepúlveda, Francisco Javier Díaz Delgadillo

Bio-inspired Optimization Methods on Graphic Processing Unit for Minimization of Complex Mathematical Functions

Although GPUs have been traditionally used only for computer graphics, a recent technique called GPGPU (General-purpose computing on graphics processing units) allows the GPUs to perform numerical computations usually handled by CPU. The advantage of using GPUs for general purpose computation is the performance speed up that can be achieved due to the parallel architecture of these devices. This paper describes the use of Bio-Inspired Optimization Methods as Particle Swarm Optimization and Genetic Algorithms on GPUs to demonstrate the performance that can be achieved using this technology with regard to use CPU primarily.

Fevrier Valdez, Patricia Melin, Oscar Castillo

Variants of Ant Colony Optimization: A Metaheuristic for Solving the Traveling Salesman Problem

Ant Colony Optimization (ACO) has been used to solve several optimization problems. However, in this paper, the variants of ACO have been applied to solve the Traveling Salesman Problem (TSP), which is used to evaluate the variants ACO as Benchmark problems. Also, we developed a graphical interface to allow the user input parameters and having as objective to reduce processing time through a parallel implementation. We are using ACO because for TSP is easily applied and understandable. In this paper we used the following variants of ACO: Max-Min Ant System (MMAS) and Ant Colony System (ACS).

Iván Chaparro, Fevrier Valdez

Intelligent Optimization Methods and Applications


Memetic Algorithm for Solving the Problem of Social Portfolio Using Outranking Model

The government institutions at all levels, foundations with private funds or private companies that support social projects receiving public funds or budget to develop its own social projects often have to select the projects to support and allocate budget to each project. The choice is difficult when the available budget is insufficient to fund all projects or proposals whose budget requests have been received, together with the above it is expected that approved projects have a significant social impact. This problem is known as the portfolio selection problem of social projects. An important factor involved in the decision to make the best portfolio, is that the objectives set out projects that are generally intangible, such as the social, scientific and human resources training. Taking into account the above factors in this paper examines the use of multi objective methods leading to a ranking of quality of all selected projects and allocates resources according to priority ranking projects until the budget is exhausted. To verify the feasibility of ranking method for the solution of problem social portfolio constructed a population memetic evolutionary algorithm, which uses local search strategies and cross adapted to the characteristic of the problem. The experimental results show that the proposed algorithm has a competitive performance compared to similar algorithms reported in the literature and on the outranking model is a feasible option to recommend a portfolio optimum, when little information and the number of projects is between 20 and 70.

Claudia G. Gómez S., Eduardo R. Fernández Gonzalez, Laura Cruz Reyes, S. Samantha Bastiani M., Gilberto Rivera Z., Victoria Ruız M.

Evolving Bin Packing Heuristic Using Micro-Differential Evolution with Indirect Representation

The development of low-level heuristics for solving instances of a problem is related to the knowledge of an expert. He needs to analyze several components from the problem instance and to think out an specialized heuristic for solving the instance. However if any inherent component to the instance gets changes, then the designed heuristic may not work as it used to do it. In this paper it is presented a novel approach to generated low-level heuristics; the proposed approach implements micro-Differential Evolution for evolving an indirect representation of the Bin Packing Problem. It was used the Hard28 instance, which is a well-known and referenced Bin Packing Problem instance. The heuristics obtained by the proposed approach were compared against the well know First-Fit heuristic, the results of packing that were gotten for each heuristic were analized by the statistic non-parametric test known as Wilcoxon Signed Rank test.

Marco Aurelio Sotelo-Figueroa, Héctor José Puga Soberanes, Juan Martín Carpio, Héctor J. Fraire Huacuja, Laura Cruz Reyes, Jorge Alberto Soria Alcaraz

Improving the Performance of Heuristic Algorithms Based on Exploratory Data Analysis

This paper promotes the application of empirical techniques of analysis within computer science in order to construct models that explain the performance of heuristic algorithms for NP-hard problems. We show the application of an experimental approach that combines exploratory data analysis and causal inference with the goal of explaining the algorithmic optimization process. The knowledge gained about problem structure, the heuristic algorithm behavior and the relations among the characteristics that define them, can be used to: a) classify instances of the problem by degree of difficulty, b) explain the performance of the algorithm for different instances c) predict the performance of the algorithm for a new instance, and d) develop new strategies of solution. As a case study we present an analysis of a state of the art genetic algorithm for the Bin Packing Problem (BPP), explaining its behavior and correcting its effectiveness of 84.89% to 95.44%.

Marcela Quiroz C., Laura Cruz-Reyes, José Torres-Jiménez, Claudia G. Gómez S., Héctor J. Fraire H., Patricia Melin

An Interactive Decision Support System Framework for Social Project Portfolio Selection

In this paper, we present the development of a Decision Support System (DSS) for the project portfolio selection problem, financed with public funds. The selection of the portfolio is a complex optimization problem with multiple subjective criteria, which are difficult to compare. The Decision Maker (DM) has to find a manageable and most preferred set among many non-dominated (efficient) solutions. There is a vast variety of techniques and software for multi-criteria decision making. However, the portfolio selection is an area that requires more and better software. An interactive “Framework” is presented; it is designed to help the DM to select the best portfolio in a flexible way. The Framework is based on the classic decision process of Simon, the SMART method and a friendly man-machine interface. The SMART method was adapted to allow the DM the discovery of his preferences and to express them on the terms of the objective weights and budget constraints. The graphic user interface assist the DM through the visualization of the impact on the change of preferences and also on the projects within a reference portfolio, this way he can make a decision or adjust the necessary changes.

Laura Cruz-Reyes, Fausto A. Balderas J., Cesar Medina T., Fernando López I., Claudia G. Gómez S., Ma. Lucila Morales R.

Constructive Algorithm for a Benchmark in Ship Stowage Planning

The efficiency of a maritime container terminal mainly depends on the process of handling containers, especially during the ships loading process. A good stowage planning facilitates these processes. This paper deals with the containership stowage problem, referred to as the Master Bay Plan Problem (MBPP). It is a NP-hard minimization problem whose goal is to find optimal plans for stowing containers into a containership with a low containership operation cost, subject to a set of structural and operational restrictions. For MBPP, data are not available for confidentiality reasons. The lack of a performance evaluation benchmark of solution algorithms for MBPP raises the need for a generation of instances. Due to this limitation, we present a generation scheme of instances for the MBPP, which is based random generation according on selected sets of parameters. The parameters are variable within certain ranges to characterize the vessel and containers; the ranges are real-life values taken from the literature. A constructive loading heuristic for stowing containers into a containership is proposed in this paper to have reference solutions. An instance set, its known-best solutions and the generator are available on-line.

Laura Cruz-Reyes, Paula Hernández H., Patricia Melin, Héctor J. Fraire H., Julio Mar O.

Small Hydroponics Garden Improved Using Cultural Algorithms

The paper discusses a research related with the innovative sense of using Decision Support System based on a Bioinspired Algorithm related with an Agribussiness and hobby too about Hydroponics, to determine the correct and adequate selection of seeds in small spaces to build scenarios of location in the future to analyze the way to improve Mexican families’ economy, this research which permits select a specific number of seeds to cultivate each one of 25 different seeds –In a time horizon of a seasonal cultivate (approximately three months)-, these seeds are evaluated from a information repository with data from another suceesful hydroponics systems. Each harvest was analyzed to built their cost-benefit during different times and scenarios and determine the viability of cultivate in the time horizon using a formal methodology based on Bioinspired Algorithms. The group of 25 seeds cultivated by “Leguizamo Povedano Cooperative” is characterized and analyzed by obtain the most representative and sucessful future scenario to determine the quantity of seeds cultivated which try to improve the limited resources and the perspectives of determine the correct selection to stablishment a signifcately life day. A case of study is presented regarding to the proposal horizons using data obtained from the Repository of the Cooperative. The intention of the present research is to apply the computational properties; in this case of established a Model of Hydropoics cultivate in a Cooperative. In addition, we analyzed the selection and location of location to cultivate a specific seed using a similarity model to locate this. The sample of study allowed analyzing the individual features of each harvest with the emulation from set matching features (commercialization, climate, in others). By means of this is possible to predict the best location to cultivte.

Alberto Ochoa-Zezzatti, Ruben Jaramillo, Sandra Bustillos, Nemesio Castillo, José Martínez, Samantha Bastiani, Victoria Ruíz

Handling of Synergy into an Algorithm for Project Portfolio Selection

Public and private organizations continuously invest on projects. With a number of candidate projects bigger than those ones that can be funded, the organization faces the problem of selecting a portfolio of projects that maximizes the expected benefits. The selection is made on the evaluation of project groups and not on the evaluation of single projects. However, there is a factor that must be taken account, since it can significantly change the evaluation of groups: synergy. This is that two or more projects are complemented in a way that generates an additional benefit to they already own individually. Redundancy, a special case of synergy, occurs when two or more projects cannot be financed simultaneously. Both features add complexity to the evaluation of project groups. This article presents an evaluation of the two most used alternatives for handling synergy, in order to incorporate it into an ant-colony metaheuristic for solving project portfolio selection.

Gilberto Rivera, Claudia G. Gómez, Eduardo R. Fernández, Laura Cruz, Oscar Castillo, Samantha S. Bastiani

Evolutionary Methods and Intelligent Computing


Practical Aspects on the Implementation of Iterative ANN Models on GPU Technology

There has been an increasing use of the graphic processing unit (GPU) in many areas including artificial neural networks (ANN) for several years. However, reported works concentrate on the application itself and not on the methodology used to implement the ANN model in the GPU. This paper presents a set of practical aspect to be considered by new GPU user in the implementation of ANN in GPUs. To illustrate the proposed aspects, the paper describes the realization of the Pulse Coupled Neural Network (PCNN), an iterative model, following these aspects and discusses the problematic of synchronization presented in this and other ANN models that is not treated in other works.

Mario I. Chacon-Murguia, Jorge A. Cardona-Soto

High Performance Architecture for NSGA-II

NSGA-II is one of the most popular algorithms for solving Multiobjective Optimization Problems. It has been used to solve different real-world optimization problems. However, NSGA-II has been criticized for its high computational cost and bad performance on applications with more than two objective functions. In this paper, we propose a high performance architecture for the NSGA-II using parallel computing, for evaluation functions and genetic operators. In the proposed architecture, the Mishra Fast Algorithm for finding the Non Dominated Set was used. We present results for five different test functions.

Josué Domínguez, Oscar Montiel, Roberto Sepúlveda, Nataly Medina

Natural Language Interfaces to Databases: An Analysis of the State of the Art

People constantly make decisions based on information, most of which is stored in databases. Accessing this information requires the use of query languages to databases such as SQL. In order to avoid the difficulty of using these languages for users who are not computing experts, Natural Language Interfaces for Databases (NLIDB) have been developed, which permit to query databases through queries formulated in natural language. Although since the 60s many NLIDBs have been developed, their performance has not been satisfactory, there still remain very difficult problems that have not been solved by NLIDB technology, and there does not yet exist a standardized method of evaluation that permits to compare the performance of different NLIDBs. This chapter presents an analysis of NLIDBs, which includes their classification, techniques, advantages, disadvantages, and a proposal for a proper evaluation of them.

Rodolfo A. Pazos R., Juan J. González B., Marco A. Aguirre L., José A. Martínez F., Héctor J. Fraire H.

A Genetic Algorithm for the Problem of Minimal Brauer Chains

Exponentiation is an important and complex task used in cryptosystems such RSA. The reduction of the number of multiplications needed during the exponentiation can significantly improve the execution time of cryptosystems. The problem of determining the minimal sequence of multiplications required for performing a modular exponentiation can be formulated using the concept of Brauer Chains.

This paper, shows a new approach to face the problem of getting Brauer Chains of minimal length by using a Genetic Algorithm (GA). The implementation details of the GA includes a representation based on the Factorial Number System (FNS), a mixture of Neighborhood Functions (NF) and a mixture of Distribution Functions (DF). We compare the proposed GA approach with another relevant solutions presented in the literature by using three benchmarks considered difficult to show that it is a viable alternative to solve the problem of getting shortest Brauer Chains.

Arturo Rodriguez-Cristerna, Jose Torres-Jimenez

Type-2 Fuzzy Grammar in Language Evolution

This paper proposes a new approach to simulating language evolution, it expands on the original work done by Lee and Zadeh on Fuzzy Grammars and introduces a Type-2 Fuzzy Grammar. Ants in an Ant Colony Optimization algorithm are given the ability of embedding a message on the pheromone using a Type-2 Fuzzy Grammar. These ants are able to gradually adopt a foreign language by adjusting the grades of membership of their grammar. Results that show the effect of uncertainty in a language are given.

Juan Paulo Alvarado-Magaña, Antonio Rodríguez-Díaz, Juan R. Castro, Oscar Castillo

Experimental Study of a New Algorithm-Design-Framework Based on Cellular Computing

In this paper the linear ordering problem with cumulative costs (LOPCC) is approached. Currently, a tabu search and a GRASP with evolutionary path-relinking have been proposed to solve it. We propose a new pseudo-parallel strategy based on cellular computing that can be applied to the design of heuristic algorithms. In this paper the proposed strategy was applied on a scatter search algorithm to show the feasibility of our approach. A series of computational experiments of the designed cellular algorithm were carried out to analyze the performance reached with the proposed approach versus its monolithic counterpart. Additionally, several parameters used were analyzed to determine their effect on the performance of the designed algorithms. We consider that the new pseudo-parallel approach used in this work can be applied to design high performance heuristic algorithms.

J. David Terán-Villanueva, Héctor Joaquín Fraire Huacuja, Juan Martín Carpio Valadez, Rodolfo A. Pazos R., Héctor José Puga Soberanes, José Antonio Martínez Flores

Restaurant Recommendations Based on a Domain Model and Fuzzy Rules

This research proposes a hybrid recommender system for restaurants that uses fuzzy inference systems together with collaborative filtering and content-based techniques, considering the expert’s experience, the ratings given by similar users and restaurant model. Content-based technique seeks to alleviate the cold-start problem, which commonly arises in collaborative filtering. The goal is to help each user to find interesting restaurants in the city. To evaluate the recommender system a data set of 50 users and 60 restaurants was tested. Was used RMSE for obtain the accuracy in the recommendations.

Xochilt Ramírez-García, Mario García-Valdez

User Modeling for Interactive Evolutionary Computation Applications Using Fuzzy Logic

Interactive evolutionary computation (IEC) is a branch of evolutionary computation where users are involved in the evolution process. In IEC systems the user generally evaluates subjective information of the population in large quantities. One of the problems in the IEC systems is not having friendly interfaces for the evaluation of mass information and this causes the user lose interest. These systems have quickly migrated to the Web by the large number of users that can be found on a voluntary basis. For these applications we can find users with different characteristics, for example, users with different level of knowledge about the application domain, different participation interest or experience in use of Web-Based IEC applications. In this paper we propose a user modeling for IEC to help tailor the user interface depending on the characteristics, preferences, interests, etc. of the user using fuzzy logic.

J. C. Romero, Mario García-Valdez

Personalization of Learning Object Sequencing and Deployment in Intelligent Learning Environments

The update and emergence of new technologies changed the traditional methods for learning in artificial environments (physical or virtual), as interactive tables, more powerful smartphones, tablet PCs, cameras that perceive depth (kinect). In this work we propose a new approach and the personalization of learning objects that we will call environmental learning object and deploy them on an intelligent learning environment in which we will use a single extension of the simple sequencing standard.

Francisco Arce, Mario García-Valdez


Weitere Informationen

Premium Partner

Neuer Inhalt

BranchenIndex Online

Die B2B-Firmensuche für Industrie und Wirtschaft: Kostenfrei in Firmenprofilen nach Lieferanten, Herstellern, Dienstleistern und Händlern recherchieren.



Product Lifecycle Management im Konzernumfeld – Herausforderungen, Lösungsansätze und Handlungsempfehlungen

Für produzierende Unternehmen hat sich Product Lifecycle Management in den letzten Jahrzehnten in wachsendem Maße zu einem strategisch wichtigen Ansatz entwickelt. Forciert durch steigende Effektivitäts- und Effizienzanforderungen stellen viele Unternehmen ihre Product Lifecycle Management-Prozesse und -Informationssysteme auf den Prüfstand. Der vorliegende Beitrag beschreibt entlang eines etablierten Analyseframeworks Herausforderungen und Lösungsansätze im Product Lifecycle Management im Konzernumfeld.
Jetzt gratis downloaden!