Fast and Effective Classification using Parallel and Multi-start PSO

Fast and Effective Classification using Parallel and Multi-start PSO

Balasaraswathi M (Department of Information Technology, SNR Sons College, Coimbatore, India) and Kalpana B (Department of Computer Science, Avinashilingam University, Coimbatore, India)
Copyright: © 2018 |Pages: 18
DOI: 10.4018/JITR.2018040102

Abstract

PSO being a swarm based algorithm, can efficiently lend itself to operate on huge data. This article presents a technique that performs classification using PSO. An initial discussion is carried out describing PSO as a classifier. Three variants of PSO are proposed here; the first variant hybridizes PSO using Simulated Annealing and the next two variants parallelizes PSO. The two parallel variants of PSO are; Parallel PSO and Multistart PSO. Parallel PSO operates by parallelizing the operation of each of the particles and Multistart PSO runs several normal versions of PSO embedded with Simulated Annealing in parallel. The multi-start version is implemented to eliminate the problem of local optima. Experiments were conducted to identify the scalability and efficiency of PSO and its variants on huge and imbalanced data.
Article Preview

Classification algorithms are usually dealt in conjunction with their application areas rather than in isolation. Only a few contributions relating to generic classification optimization are available. This section presents some of the PSO based classifiers and their utilization scenarios.

A generic classification technique using PSO as the base classifier was presented by Falco et al. in (De Falco, Della Cioppa, &Tarantino, 2007). This technique provides three different fitness functions to be used as the base for classification. It discusses the base of PSO and the operational scenario of applying PSO on a classification problem. Results obtained from using PSO on different fitness functions were compared against statistical classifiers and PSO was found to be efficient in classifying the results. PSO, being the new comer in the classification domain, contains very few contributions to its credit. A comparison between Genetic algorithm, tree induction algorithm and PSO in the area of classification was presented in (Sousa, Silva, & Neves, 2004). A major use of PSO was found in the image classification domain. One such technique that uses PSO for classifying images was presented by Omran et al. in (Brits, Engelbrecht, & Van den Bergh, 2004). Another similar image classification system was presented by Xue et al. (2014). This technique uses SVM as the base classifier and PSO as an optimizer for SVM.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 12: 4 Issues (2019): 1 Released, 3 Forthcoming
Volume 11: 4 Issues (2018)
Volume 10: 4 Issues (2017)
Volume 9: 4 Issues (2016)
Volume 8: 4 Issues (2015)
Volume 7: 4 Issues (2014)
Volume 6: 4 Issues (2013)
Volume 5: 4 Issues (2012)
Volume 4: 4 Issues (2011)
Volume 3: 4 Issues (2010)
Volume 2: 4 Issues (2009)
Volume 1: 4 Issues (2008)
View Complete Journal Contents Listing