Binary Biogeography-Based Optimization Applied to Gene Selection for Cancer Classification Using Artificial Neural Network
Dilwar Hussain Mazumder, Ramachandran Veilumuthu · 2018
In cancer classification, selection of genes that highly contribute to the classification process becomes essential due to the problem of 'curse of dimensionality' associated with microarray based gene expression data. Biogeography-Based Optimization (BBO) is a population based evolutionary computation technique successfully applied to many application domains and proved to deliver optimal solutions. This work proposes a gene selection method named as BBBOFS, by applying Binary Biogeography-Based Optimization (BBBO). The selected genes are used for cancer classification using the Artificial Neural Network (ANN) classifier. The proposed method is validated through experiments on standard gene expression dataset benchmarks. Results demonstrate that the proposed method is better than the related works from literature in terms of classification accuracy and selected gene count.