Hybrid binary Coral Reefs Optimization algorithm with Simulated Annealing for Feature Selection in high-dimensional biomedical datasets

The last decades have witnessed accumulation in biomedical data. Though they can be analyzed to enhance assessment of at-risk patients and improve the diagnosis, a major challenge associated with biomedical data analysis is the so-called “curse of dimensionality”. For the issue, an improved Coral Re...

Full description

Saved in:
Bibliographic Details
Published in:Chemometrics and intelligent laboratory systems 2019-01, Vol.184, p.102-111
Main Authors: Yan, Chaokun, Ma, Jingjing, Luo, Huimin, Patel, Ashutosh
Format: Article
Language:English
Subjects:
Biomedical dataset
Coral reefs optimization
Feature selection
Simulated annealing
Tournament selection
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The last decades have witnessed accumulation in biomedical data. Though they can be analyzed to enhance assessment of at-risk patients and improve the diagnosis, a major challenge associated with biomedical data analysis is the so-called “curse of dimensionality”. For the issue, an improved Coral Reefs Optimization algorithm for selecting the best feature subsets has been proposed. Tournament selection strategy is adopted to increase the diversity of initial population individuals. The KNN classifier is used to evaluate the classification accuracy. Experimental results on thirteen public medical datasets show proposed BCROSAT outperforms other state-of-the-art methods. •A novel framework based on an improved Coral Reefs Optimization is applied to feature selection for biomedical data.•Tournament selection strategy is employed to produce initial coral reef population.•Simulated Annealing (SA) is combined with CRO to improve the search performance of the original CRO algorithm.
ISSN:0169-7439
1873-3239
DOI:10.1016/j.chemolab.2018.11.010