Feature Subset Selection Based on Bio-Inspired Algorithms *

• DOI: 10.6688/jise.2011.27.5.10
• Publication Date: 2011-09-01
• Authors: Chulmin Yun; Byonghwa Oh; Jihoon Yang; Jongho Nang
• Keywords: Feature (linguistics); Relevance

Many feature subset selection algorithms have been proposed and discussed for years. However, the problem of finding optimal from full data still re- mains to be a difficult problem. In this paper, we propose novel methods find rele- vant by using biologically-inspired such as Genetic Algorithm Particle Swarm Optimization. We also variant approach considering significance each feature. verified performance experiments with various real-world datasets. Our based on produced better than other in terms classification accuracy relevance. particular, modified method demonstrated even more improved performance.