Towards optimal feature and classifier for gene expression classification of cancer

Recently, demand on the tools to efficiently analyze biological genomic information has been on the rise. In this paper, we attempt to explore the optimal features and classifiers through a comparative study with the most promising feature selection methods and machine learning classifiers. In order to predict the cancer class, the gene information from patient’s marrow expressed by DNA microarray, who has either the acute myeloid leukemia or acute lymphoblastic leukemia. Pearson and Spearman’s correlation, Euclidean distance, cosine coefficient, information gain, mutual information and signal to noise ratio have been used for feature selection. Backpropagation neural network, self-organizing map, structure adaptive self-organizing map, support vector machine, inductive decision tree and k-nearest neighbor have been used for classification. Experimental results indicate that backpropagation neural network with Pearson’s correlation coefficients is the best method, obtaining 97.1% of recognition rate on the test data.