Classification Algorithm Analysis for Breast Cancer

Abstract

Breast cancer in women is a type of disease that is the main cause of death in women according to world breast cancer data. Therefore, early detection of breasts is needed significantly to improve life. If a woman has been identified, then rehabilitation and treatment on an incentive basis are needed to reduce the worse. This study used a dataset collected by the University of Wisconsin Hospitals, Madison (https://atapdata.ai/). This research conducted experiments using several data mining classification strategies to predict breast cancer using machine learning algorithms. The Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Naive Bayes, Random Forest, Decision Tree, Deep Learning (H2O), and Neural Network are used to classify algorithms. From these algorithms’ classification, we compare accuracy, best classification, and compare algorithm performance with curve ROC (RapidMiner Studio Core) to see which performance algorithm has the best quality for classification. From the analysis results, the deep learning algorithm with Tanh and Exprectifier activation function has a good accuracy of 93.14%, and the best classification with 89.62%. In addition, deep learning has found the best quality from the ROC curve results on the dataset used in this research.

Performance enhancement of machine learning algorithm for breast cancer diagnosis using hyperparameter optimization

“…On the other hand, both SVM and MLP obtained 96.50% accuracy without PSO. To classify BC, Sukmandhani et al [18] analyzed the classification performance of SVM, NB, k-NN, RF, DT, NN, and H2O models. Among ML models, RF obtained the highest accuracy which was 92.26%, and SVM, NB, KNN, and DT got 88.59%, 90.52%, 88.93%, and 90.50% accuracy, respectively.…”