SMO, IBK and Logistic Classifiers for Classification of Breast Cancer

Abstract

Breast cancer has become common cancer among various ages of women in world. The symptoms of breast cancer include the lumps, change in shape, change in color of skin, liquid oozing out from nipple. The occurrence of breast cancer is increasing every year by year and increased incidence is not surprising since there has been increase in numbers of women with high breast cancer risk factors, including lower age of menarche, late age of first pregnancy, fewer pregnancies, shorter or no periods of breastfeeding, later menopause, prolactin hormone levels, use of birth control pills and radiation exposure in youth. In order to reduce the mortality of breast cancer effective early classification detection techniques are required for long term survival of the women. Classification is one of the most commonly used tool analyze and classify the data. This paper analyzes the classifier algorithms such as SMO, Logistic and MLP for seer breast cancer dataset using WEKA software. The performance of the classifiers are evaluated against the parameters like classification accuracy, RMSE, TP Rate, FP Rate, Precision, Recall, F-Measure, ROC, Specificity, Sensitivity and so on. MLP Classifier is superior to SMO and Logistic and it has the highest classification accuracy of 93.5%.