Breast cancer prediction based on multiple machine learning algorithms

Abstract

Breast cancer is one of the common cancers, and its timely and accurate prediction is necessary. In this study, a variety of breast cancer prediction algorithms implemented in Python are studied, compared, and optimized horizontally and vertically. In this study, the problem of breast cancer prediction is deeply studied. By comparing algorithms such as logistic regression, decision tree, random forest, K-Nearest Neighbor (KNN) and support vector Machine (SVM), and introducing two optimization methods, to explore the effective ways to improve the accuracy of breast cancer prediction. It was found that of the five machine learning algorithms provided by the tested sklearn library, logistic regression, random forest and support Vector Machine (SVM) performed well in this breast cancer prediction dataset application. The method of slightly increasing the fitting was calculated and used, the logistic regression statistical fitting method was used and the data was predicted again to obtain better prediction results. Finally, the prediction accuracy of 98.83% was achieved by the optimization method. This provides important guidance for decision makers in the selection of appropriate breast cancer prediction algorithms, which provides stronger support for the early diagnosis and treatment of breast cancer in the future.