A COMPARATIVE STUDY ON THE APPLICATION OF BINARY PARTICLE SWARM OPTIMIZATION AND BINARY GRAVITATIONAL SEARCH ALGORITHM IN FEATURE SELECTION FOR AUTOMATIC CLASSIFICATION OF BRAIN TUMOR MRI

Authors

  • M. A. Majid
  • A. F. Z. Abidin
  • N. D. K. Anuar
  • K. A. Kadiran
  • M. S. Karis
  • Z. M. Yusoff
  • N. H. K. Anuar
  • Z. I. Rizman

DOI:

https://doi.org/10.4314/jfas.v10i2s.35

Keywords:

brain tumor, gray level co-occurrence matrix, nearest neighbor, binary particle swarm optimization, binary gravitational search algorithm.

Abstract

Researches on automatic classification for brain tumor had been done extensively, yet there is still room for improvement. Many approaches have been focused on image segmentation and classifier algorithm, yet little number of researches done on feature selection. This paper presents a study on the applications of two popular Swarm Intelligence algorithms: Binary Particle Swarm Optimization and Binary Gravitational Search Algorithm for optimizing feature selection of Gray-Level Co-occurrence Matrix. The classifier that is used in this paper is k-Nearest Neighbor. Benchmarking is done by comparing both swarm intelligence algorithms mentioned. The result indicates Binary Particle Swarm Optimization performs better compared to Binary Gravitational Search Algorithm.

Downloads

Published

2018-02-01

Issue

Section

Research Articles