A Hybrid GA/kNN/SVM Algorithm for Classification of data

Dr. Subash Chandra Bose, J and Suresh Babu, Changalasetty and Ahmed Said Badawy, Dr and Wade Ghribi, Dr and Jamel Baili, Dr and Harun Bangali, Mr (2016) A Hybrid GA/kNN/SVM Algorithm for Classification of data. Biohouse Journal of Computer Science, 2 (2). pp. 5-11. ISSN ISSN 2379-1500

[img] PDF
BJCS-V2-I2-3.pdf

Download (315kB)
Official URL: http://biohouse.us/docs/BJCS-V2-I2-3.pdf

Abstract

This paper proposes an effective comparison of different algorithms for classification namely Genetic Algorithm, K-Nearest Neighbor (kNN) and Support Vector Machines (SVM) techniques. The goal of the comparison is to compare the effects of the classification rules from data. The algorithm is stimulated by the behavior of classification. Individual data are selected based on how well they fit to the environment. In GP, the entity that reflects this degree of adaptation is the fitness function. The programs that better solve the problem at hand will receive a better fitness value, and will consequently have a better chance of being selected. The datasets that are considered can be any data involving choice of a fitness function, and an evaluation method depends on the problem metrics given to the simulation environment based on parameter values. SVM and KNN classifiers are compared and applied their performance using MATLAB simulation environment. Keywords— K nearest neighboring, Genetic Algorithm, Support Vector Machine, Micro array, Classification.

Item Type: Article
Subjects: Computer Engineering
Computer Sciences
Divisions: College of Computer Sciences > Computer Engineering
Depositing User: subash jaganathan
Date Deposited: 16 Mar 2017 08:31
Last Modified: 16 Mar 2017 08:31
URI: http://eprints.kku.edu.sa/id/eprint/533

Actions (login required)

View Item View Item