Forward feature extraction from imbalanced microarray datasets using wrapper based incremental genetic algorithm. (3rd November 2020)

• Record Type: Journal Article
• Authors: Priya, R. Devi; Sivaraj, R.; Anitha, N.; Devisurya, V.
• Is Part Of: International journal of bio-inspired computation. Volume 16:Number 3(2020)
• Page Start: 171
• Page End: 180
• Publication Date: 2020-11-03
• Subjects: incremental genetic algorithm -- IGA -- imbalanced dataset -- bootstrap sampling -- forward feature extraction -- adaptive mutation; Biologically-inspired computing -- Periodicals; Computational biology -- Periodicals; 572.0285
• Languages: English
• ISSNs: 1758-0366

Abstract

Learning from imbalanced datasets is a critical challenge confronting researchers. Unequal distribution of classes in the imbalanced datasets lead to biased classification especially in microarray gene expression analysis. Since all features in the dataset will not contribute to the analysis, only prominent and significant features need to be identified. The paper addresses both these issues by proposing wrapper based incremental genetic algorithm (IGA) which incrementally evaluates and adds attributes into the genetic algorithm process rather than evaluation of all attributes thereby reducing the computational complexity and number of features used and improving the measures like classification accuracy, GMean, F1 measure, precision and recall. The experiments are conducted on 8 microarray gene expression datasets and the results show that performance of IGA is encouraging and superior to existing methods that are compared.