Automated Breast Cancer Diagnosis Based on Machine Learning Algorithm

Abstract – Breast Cancer classification is becoming more important with the increasing demand of automated applications especially interactive applications. It can be used to improve the performance of classifiers like Logistic Regression, Decision Tree, Random Forest, SVC etc. This study is based on learning genetic patterns of patients with breast tumors and machine learning algorithms that aim to demonstrate a system to accurately differentiate between benign and malignant breast tumors. The aim of this study was to optimize different algorithm. In this context, we applied the genetic programming technique to select the best features and perfect parameter values of the machine learning classifiers. The performance of the proposed method was based on accuracy, precision and the roc curves. The present report prepared by us proves that genetic programming can automatically find the best model by combining feature preprocessing methods and classifier algorithms by reducing False Positive rate. In this paper, there were two challenges to automate the breast cancer diagnosis: (i) determining which model best classifies the data and (ii) how to automatically design and adjust the parameters of the machine learning model. We have summarized the experimental studies and the obtained results, and lastly presented the main conclusion.