Identifying cancer biomarkers from leukemia data using feature selection and supervised learning

The disease Leukemia are continuously increasing among the people. The cause of leukemia is unknown but several factors, however are associated with the development of leukemia that are exposure to ionising radiation, exposure to benzene in rubber industry workers, cytotoxic drug particularly alkylating agent exposure, genetic disorder like down syndrome and immunological deficiency states. There are certain medical diagnosis that can reduce the burden on doctors. This paper explores some algorithms for the classification of Leukemia patients dataset. The algorithms considered here are Naive Bayes (NB), K Nearest Neighbour (KNN) and Support Vector Machines (SVMs).