Medical Image Prediction for Diagnosis of Breast Cancer Disease Comparing the Machine Learning Algorithms: SVM, KNN, Logistic Regression, Random Forest, and Decision Tree to Measure Accuracy

Abstract

Machine learning using data mining techniques are used rapidly in medical research to predict the disease diagnosis. The aim of this study is to evaluate the performance of SVM, KNN, Logistic Regression, Random Forest, and Decision Tree. Materials and Methods: A Total of 569 samples are collected from the UCI Machine Learning Laboratory. The samples are divided into benign and malignant cells using groups like SVM, KNN, Decision Tree, Random Forest, and Logistic Regression to compare the performance of benign and malignant cells. The required samples for this analysis are done by G power calculation. Minimum power of analysis is fixed as 0.8 and maximum accepted error is fixed as 0.5. Results: Logistic Regression prediction appears to be better accuracy of 95% than SVM, KNN, Decision Tree, and Random Forest of 92%, 90%, 85%, 91%. Significance of this proposed system is likely to be 0.55. Conclusion: In this study, it is found that Logistic Regression appears to be better than the SVM, KNN, Decision Tree, and Random Forest in breast cancer detection using the Wisconsin dataset.