Breast Cancer Classification: Using Hybrid Technique (Association Rules and Deep Neural Network)

Abstract

Early detection of any disease is very important since it aids curable with a few of effort. A lot of people fail to detection their disease before it be chronic. This causes an increase in mortality about the world. One of these diseases is breast cancer that can be cured when identified in the early stages before it spreads throughout the body. Develop techniques that can aide physicians to get accurate diagnosis is significantly important in early detection for this disease. The goal is design a hybrid approach (class association rules and deep neural network). In this paper, we design efficient methodology for classifying breast cancer using hybrid approach techniques. Where used a CARs to discover all the interesting relationship in a large database, while the DNN is used for classification purpose. In this study, use Wisconsin Breast cancer dataset from UCI machine learning repository to evaluate the performance of the proposed system. The experiment show that proposed system achieves good results, with high accuracy of 100% and less mean square error rate 0.0002.