Using an Adaptive Linear Support Vector Machine Algorithm for Predicting the Breast Cancer

Abstract

Breast cancer is the most common type of cancer and a significant contributor to the high death rates among women. The death rate increases when this condition is manually diagnosed causing delay of cancer detection since it takes several hours and requires the availability of specialists. Therefore, an automated breast cancer diagnosis has been suggested to speed up detection and stop the disease from spreading. Over the years, machine learning classification algorithms have been used to predict breast cancer. In previous studies, one of the most used algorithms is the Support Vector Machine (SVM). However, these studies have inconsistent results. This work investigates the impact of the features' selection, hyperparameters of SVM, and the mechanism of splitting data on the algorithm performance, thus, building an SVM, as a single machine learning model, that achieves a higher accuracy. The Wisconsin dataset was used to train and test this model. The experimental results showed that the performance of the model was affected by the features' selection, hyperparameters, and the mechanism of splitting data and random state values in terms of the best results and the average of the top three results. The comparison results revealed the superiority of the proposed method over the other state-of-the-art methods.