A new classifier for breast cancer detection based on Naïve Bayesian Murat Karabatak Fırat University, Department of Software Engineering, 23119 Elazig, Turkey article info Article history: Received 19 March 2015 Accepted 20 April 2015 Available online 6 May 2015 Keywords: Breast cancer detection NB classifier Performance evaluation tests Weighted NB classifier abstract Breast cancer is known as the most common invasive cancer type among women and automatic breast cancer detection systems are in demand. Thus, various machine learning and pattern recognition techniques have been proposed to detect breast cancer. One of these techniques is the Bayes classifier. Naïve Bayesian (NB) is known to be a simple classifier, which is based on the Bayes theorem. There have been so many applications used in literature. In this paper, a new NB (weighted NB) classifier was proposed and its application on breast cancer detection was presented. Several experiments were conducted to evaluate the performance of the weighted NB on the breast cancer database. The exper- iments were realized with 5-fold cross validation test. Moreover, various performance evaluation techniques namely sensitivity, specificity and accuracy are considered. According to the experiments, the weighted NB obtained the following evaluation values. The calculated sensitivity, specificity and the accuracy values are 99.11%, 98.25%, and 98.54% respectively. Moreover, a comparison with the existing methods in the literature was presented. As a result, the performance of weighted NB is better than regular NB and many other existing methods. Ó 2015 Elsevier Ltd. All rights reserved. 1. Introduction Breast cancer is the leading cancer type in females and is the fifth most common cause of cancer death [1]. In 2013, American Cancer Society (ACS) reported that 39,620 women and 410 men, a total of 40,030 people, passed away due to breast cancer. According to the physicians, an abnormal multiplication of cells in the breast tissue causes breast cancer [2]. Mammography is a traditional method that has been used to detect the breast cancer [3]. Interpreting mammography necessitates highly skilled radiologists because in literature, radiologists report various interpretations for the same mammography [4]. Masses and micro-calcifications are considered two common types of abnormalities in mammograms [5]. In breast tissue, masses look like lesions, lumps, or protuber- ances in the breast and micro-calcifications are calcium deposits in the breast appearing in clusters or individual spots. Individual micro-calcifications have different sizes ranging from twenty to hundreds of microns in diameter. Micro-calcifications can be differentiated due to their higher contrast than masses in mammograms thus, mak- ing detection and diagnosis of masses challenging [6].A widely used diagnostic tool for breast cancer is fine needle aspiration cytology (FNAC). FNAC’s correct classification rate is about 90% and there is a dire need to have improved classification/detection systems are in demand. Recently, artificial intelligence and machine learning techniques have been widely applied in detection/recognition of the breast cancer [7,8]. The common aim of these methods is to classify the patients to either a ‘benign’ group or a ‘ma- lignant’ group. Benign group corresponds to the patients http://dx.doi.org/10.1016/j.measurement.2015.04.028 0263-2241/Ó 2015 Elsevier Ltd. All rights reserved. E-mail address: muratkar@hotmail.com Measurement 72 (2015) 32–36 Contents lists available at ScienceDirect Measurement journal homepage: www.elsevier.com/locate/measurement