Skip to main content

Über dieses Buch

In this book a new model for data classification was developed. This new model is based on the competitive neural network Learning Vector Quantization (LVQ) and type-2 fuzzy logic. This computational model consists of the hybridization of the aforementioned techniques, using a fuzzy logic system within the competitive layer of the LVQ network to determine the shortest distance between a centroid and an input vector. This new model is based on a modular LVQ architecture to further improve its performance on complex classification problems. It also implements a data-similarity process for preprocessing the datasets, in order to build dynamic architectures, having the classes with the highest degree of similarity in different modules. Some architectures were developed in order to work mainly with two datasets, an arrhythmia dataset (using ECG signals) for classifying 15 different types of arrhythmias, and a satellite images segments dataset used for classifying six different types of soil. Both datasets show interesting features that makes them interesting for testing new classification methods.



Chapter 1. Introduction

A classificationClassification problem consists in categorizing an object based on certain attributes, with the aim of identifying to which classClass it belongs to. For instance, a fruit could be classified based on its size, color, or shape; the same way as an automobile, a flower, an animal, among others. All these objects have their own attributes, and which attributes are considered for classifying an object (or event) will depend on the problem to work with. For example, a heart disease could be classified using data obtained from a Holter device, a tumor or a cancer cell could be classified based on the data of an image.

Jonathan Amezcua, Patricia Melin, Oscar Castillo

Chapter 2. Theory and Background

Computer science embraces a variety of different areas such as Computer Graphics, Computational Complexity, Computer Cryptography, Computational Intelligence, among others.

Jonathan Amezcua, Patricia Melin, Oscar Castillo

Chapter 3. Problem Statement

Learning Vector QuantizationLearning Vector Quantization (LVQ) (LVQ) is an algorithm widely used for solving classificationClassification problems. Some works include [1] where the algorithm was used for classifying faulty LEDs, in Fallah et al. (Proceedings of International MultiConference of Engineers and Computer Scientists, IMECS 2010. Hong Kong [2]) LVQ was used for iris recognition and classification using an artificial vision system. In addition in recent work LVQ has been also used for arrhythmia classification with a modular architecture (Melin et al. in Informatics and Computer Science Intelligent Systems Applications. Information Sciences 279:483–497, 2014 [3]).

Jonathan Amezcua, Patricia Melin, Oscar Castillo

Chapter 4. Proposed Classification Method

In this section the proposed Fuzz LVQLearning Vector Quantization (LVQ) method for classificationClassification is described. This is based on LVQ neural networksNeural networks and fuzzy systemsFuzzy system for the classification of arrhythmias, and different types of soil based on satellite images segments. The fuzzy approach was implemented in the competitive layer of the LVQ algorithm, letting the fuzzy system to determine which of the cluster centers is the nearest to an input vector, and then moving the cluster center either towards or away from the input vector. Fuzz LVQ modular architectures were developed in order to test two datasets.

Jonathan Amezcua, Patricia Melin, Oscar Castillo

Chapter 5. Simulation Results

In this section, the obtained results are discussed. For all simulations (MIT-BIH arrhythmia dataset and Satellite Images dataset), 70% of the datasets was considered for training the models, and the 30% was used for testing. The testing sets were chosen randomly from the whole dataset, this means, the testing sets were always composed by different vectors for testing the models.

Jonathan Amezcua, Patricia Melin, Oscar Castillo

Chapter 6. Conclusions

In this book the Fuzz LVQLearning Vector Quantization (LVQ) method for classificationClassification tasks is presented. This new method is based on the hybridization of artificial neural networksNeural networks with the LVQ algorithm and type-2 fuzzy logic. Classification of information can be a complicated task. In general terms, for working with LVQ networks, and some other classification methods, it is important to thoroughly analyze the information and determine the most representative attributes. This is helpful by itself and avoids an overload of information for the method.

Jonathan Amezcua, Patricia Melin, Oscar Castillo


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.



Best Practices für die Mitarbeiter-Partizipation in der Produktentwicklung

Unternehmen haben das Innovationspotenzial der eigenen Mitarbeiter auch außerhalb der F&E-Abteilung erkannt. Viele Initiativen zur Partizipation scheitern in der Praxis jedoch häufig. Lesen Sie hier  - basierend auf einer qualitativ-explorativen Expertenstudie - mehr über die wesentlichen Problemfelder der mitarbeiterzentrierten Produktentwicklung und profitieren Sie von konkreten Handlungsempfehlungen aus der Praxis.
Jetzt gratis downloaden!