Classification of cancer data using support vector machines with features selection method based on global artificial bee colony

Cancer is the second leading cause of death globally. Various methods have been proposed to deal with cancer disease. One of this method is cancer classification with features selection based on microarray data. In this study, we proposed Global Artificial Bee Colony (GABC) - Support Vector Machines (SVM) for classification of cancer. GABC is advance version of standard ABC, has a global ability to find global solution result, has fewer control parameters and easy to implement. GABC is applied in order to select the most informative features for cancer classification. The experiments to test the accuracy of classification were conducted on two binary microarray datasets: lung cancer dataset is obtained from UJN and breast cancer dataset is obtained from NCBI. The results show that classification accuracy on breast cancer dataset is 96.4286 % and 99.9999 % on lung cancer dataset, compared with classification without features selection on breast dataset is 66.6667 % and lung dataset is 82.9819 %. Therefore,...