Performance Comparison of Three Classifiers for the Classification of Breast Cancer Dataset

Breast Cancer is one of the threatening issues for women's existence nowadays. It is increasing in our society due to pursuing modern/western cultures and careless in food and living habits. It has some syndromes and based on those syndromes we can easily identify whether a patient has breast cancer or not. Support Vector Machine (SVM), Artificial Neural Network (ANN) and Naïve Bayes Algorithms are very popular and powerful supervised learning algorithms to classify an unknown label/result. We select a dataset from WBCD (Wisconsin Breast Cancer Diagnosis) which contains 9 attributes column and 1 class column. The attribute columns are the causes and the class column is the result of the attribute columns. In this paper, we trained different parts of SVM, ANN and Naïve Bayes based on a particular training dataset (WBCD). Based on the highest accuracy, we voted the best model from the described models in this paper and selected it to use in the future for the client dataset (clinical data). The best model is Linear SVM for the WBCD dataset and accuracy is 96.72%.