Comparison of microarray breast cancer classification using support vector machine and logistic regression with LASSO and boruta feature selection

Nursabillilah Mohd Ali (Universiti Teknikal Malaysia Melaka)
Nor Azlina Ab Aziz (Multimedia University)
Rosli Besar (Multimedia University)

Indonesian Journal of Electrical Engineering and Computer Science
Vol 20, No 2: November 2020

Breast cancer is the most frequent cancer diagnosis amongst women worldwide. Despite the advancement of medical diagnostic and prognostic tools for early detection and treatment of breast cancer patients, research on development of better and more reliable tools is still actively conducted globally. The breast cancer classification is significantly important in ensuring reliable diagnostic system. Preliminary research on the usage of machine learning classifier and feature selection method for breast cancer classification is conducted here. Two feature selection methods namely Boruta and LASSO and SVM and LR classifier are studied. A breast cancer dataset from GEO web is adopted in this study. The findings show that LASSO with LR gives the best accuracy using this dataset.