Hybrid Genetic-Fuzzy Algorithm for Variable Selection in Spectroscopy

This paper presents a hybrid multi-objective genetic fuzzy algorithm for the variable-selection problem in spectroscopy. The problem formulation considers three fitness functions related to linear equations system stability. These fitness functions are models with fuzzy sets that evaluate the fitnes...

Full description

Saved in:
Bibliographic Details
Published inArtificial Intelligence and Soft Computing pp. 24 - 35
Main Authors de Lima, Telma Woerle, da Silva Soares, Anderson, Coelho, Clarimar José, Salvini, Rogério Lopes, Laureano, Gustavo Teodoro
Format Book Chapter
LanguageEnglish
Published Berlin, Heidelberg Springer Berlin Heidelberg 2013
SeriesLecture Notes in Computer Science
Subjects
Fuzzy logic
Multi-objective Genetic Algorithm
Variable Selection
Online AccessGet full text
ISBN3642386091
9783642386091
ISSN0302-9743
1611-3349
DOI10.1007/978-3-642-38610-7_3

Cover

More Information
Summary:This paper presents a hybrid multi-objective genetic fuzzy algorithm for the variable-selection problem in spectroscopy. The problem formulation considers three fitness functions related to linear equations system stability. These fitness functions are models with fuzzy sets that evaluate the fitness solution for pick out the best to crossover. The population diversity is obtained applying the crowding distance method. The study shows that the selection by a fuzzy decision has better results than the selection by non-domination in problems where the fitness weighing is more proper than no-domination solutions.
ISBN:3642386091
9783642386091
ISSN:0302-9743
1611-3349
DOI:10.1007/978-3-642-38610-7_3