Applying Grey Wolf Optimizer-based decision tree classifer for cancer classification on gene expression data

One of the most important research topics in bioinformatics is classification of cancer and in this topic, one of major concepts is using of microarray data. There are lots of methods ranging from analytical and computational intelligence methods for using gene expression data such as Back Propagation Neural Network (BPNN), Self-Organizing Map (SOM), Support Vector Machine (SVM) and Particle Swarm Optimization (PSO). In this paper, a hybrid method of Grey Wolf Optimizer (GWO) combined with decision tree as a classifier for selecting a small number of useful genes from the lots of genes to identify cancer is proposed. Proposed method and other famous classifiers such as BPNN, SOM, SVM, C4.5 and PSOC4.5 applied to experiments on 10 gene expression cancer datasets and results show proposed method outperforms others.