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

摘要

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-theart methods.