Accuracy Assessment of Several Machine Learning Algorithms for Breast Cancer Diagnosis

Abstract

A number of critical metrics, such as confusion matrix, precision, recall, F1-score, support, and accuracy, were used to evaluate and rank several machine learning classification methods. Linear discriminant analysis, logistic regression, decision tree classification, k-nearest neighbors, gaussian naive bayes, support vector machine, and random forest were only few of the statistical methods used to evaluate the Breast Cancer Wisconsin Diagnostic dataset. The accuracy rate of the Logistic Regression classifier is higher than that of its competitors. The assignment was accomplished in the Anaconda environment using the Python programming language and the Scikit-learn package.