An Efficient Multi-Layer Perceptron Neural Network-Based Breast Cancer Prediction
Saravana Kumar N. M., S. Tamilselvi, K. Hariprasath, N. Kaviyavarshini, A. Kavinya · Advances in medical technologies and clinical practice book series · 2022
Cancer is the most deadly disease for human beings across the world due to the adoption of new food habits and pollution. Breast cancer is becoming a common disease for women. After years of research a better analysis and possessing a higher prediction of any kind of cancer, it is still a very imperative contribution in healthcare. In the literature, several attempts were made deploying machine learning methods yielding a significant prediction of cancer. The goal of this study is on machine learning techniques that can enable prediction of breast cancer from the attributes with more accuracy. In this study, Wisconsin Breast Cancer (WBC) dataset has been used to compute machine and deep learning algorithm. This research focuses on logistic regression, decision tree, and random forest and a novel deep learning adoption, multi-layer perceptron. The performances of these algorithms have been correlated with each other using accuracy percentage. Also, sensitivity and specificity are computed as evaluation metrics. The multi-layer perceptron (MLP) gives 97% accuracy.