Using Feature Selection Techniques to Improve the Accuracy of Breast Cancer Classification

Classification is a data mining process that aims to divide data into classes to facilitate decision-making; it is therefore an important task in medical field. In this paper we will try to improve the accuracy of the classification of six machines learning algorithms: Bayes Network (BN), Support Vector Machine (SVM), k-nearest neighbors algorithm (Knn), Artificial Neural Network (ANN), Decision Tree (C4.5) and Logistic Regression using feature selection techniques, for breast cancer classification and diagnosis. We examined those methods of classification and techniques of feature selection in WEKA Tool (The Waikato Environment for Knowledge Analysis) using two databases, Wisconsin breast cancer datasets original (WBC) and diagnostic (WBCD) available in UCI machine learning repository.