Sequential Combination of Two Classifier Algorithms for Binary Classification to Improve the Accuracy
Sornxayya Phetlasy, Satoshi Ohzahata, Celimuge Wu, Toshihiko Kato · 2015
Binary classification is a process of classifying the elements of a data set into two groups on the basis of a classification rule. It is useful and widely applied in many fields: Information Technology, Business, Medical Diagnosis, Finance, and so on. The problems of the previous works do not specify clearly which classifier utilizes to minimize which type of false, False Positive (FP) or False Negative (FN), because they are tradeoffs. In this study, we propose a hybrid method for data classification with two different classifier algorithms. The first classifier responds to reduce FN, and the second classifier is in charge of reducing FP. Our experiments utilize the data set of breast cancer, Wisconsin Breast Cancer (WBC) which is popular data set among the researchers for breast cancer diagnosis. The results show that the proposed method improves accuracy.