An Island Grouping Genetic Algorithm for Fuzzy Partitioning Problems

Abstract

This paper presents a novel fuzzy clustering technique based on grouping genetic algorithms (GGAs), which are a class of evolutionary algorithms especially modified to tackle grouping problems. Our approach hinges on a GGA devised for fuzzy clustering by means of a novel encoding of individuals (containing elements and clusters sections), a new fitness function (a superior modification of the Davies Bouldin index), specially tailored crossover and mutation operators, and the use of a scheme based on a local search and a parallelization process, inspired from an island-based model of evolution. The overall performance of our approach has been assessed over a number of synthetic and real fuzzy clustering problems with different objective functions and distance measures, from which it is concluded that the proposed approach shows excellent performance in all cases.

An Accelerated Optimization Approach for Finding Diversified Industrial Group Stock Portfolios with Natural Group Detection

“…They also discussed an island model to improve efficiency. Based on a GGA, Salcedo-Sanz et al proposed a fuzzy clustering technique using a modified version of the DBI [18]. Chen et al used Ward's method with hierarchical clustering to determine clustering quality and select a portfolio based on equity mutual funds [34].…”

“…The method encoded a DGSP into a chromosome comprising four parts, namely the active stock part, grouping part, stock part, and stock portfolio part. Then, each chromosome was evaluated not only by the portfolio satisfaction (PS) and group balance (GB), like in prior work [17], but also by the clustering quality (CQ), for example, the Davis-Bouldin index (DBI) [18], to find the DGSP and its suitable stock group size. Furthermore, the temporary chromosome (TC), which aimed to minimize the number of combinations required, was designed to speed up the evolutionary time.…”

Bio-inspired computation: Where we stand and what's next

“…In all the tested problems the proposed dEA-based discretizer performed best, attaining superior accuracy rates. Unsupervised classification over large-scale datasets has also been tackled by adopting dEAs, such as [221], where an island genetic algorithm is proposed for fuzzy partition problems, or [222], where dEAs are applied to improve a k-Means clustering algorithm. There has been also active research around more practical versions of dEAs, in areas such as large-scale optimization [223,224,225], Electromagnetism [226], Computational Fluid Mechanics [227], energy planning [228] or neural networks training [229].…”

K-Means Cloning: Adaptive Spherical K-Means Clustering

“…Such methods try to find global optimal configurations of data clusters within reasonable processing times. Different types of metaheuristics have been used to solve different data clustering problems such as genetic algorithms [45][46][47][48][49][50][51], simulated annealing [18,52,53], tabu search [54,55], particle swarm optimization [54,[56][57][58][59], ant colony optimization [60][61][62][63], harmony search algorithm [64,65], and firefly algorithm [66,67].…”