Evaluating the Performance of Supervised Machine Learning Algorithms for Predicting Multiple Diseases: A Comparative Study

Machine learning has become one of the most popular and widely used techniques in various industries, especially in the healthcare sector. This paper focuses on examining the performance analysis of several machine learning algorithms and conducting a comparative analysis to determine which algorithm performs best in predicting six different diseases, such as cardiovascular, heart, stroke, Alzheimer, breast, and lung cancer, with various metrics. The Multi Disease Dataset was obtained from the UCI Machine Learning Repository and contains various instances and attributes. The proposed model involves two stages of data preprocessing, including missing value replacement and label encoding. Additionally, Six diverse supervised machine learning classification algorithms, including Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Logistic Regression (LR), Naive Bayes (NB), Random Forest (RF), and Decision Tree (DT), were analyzed. Also the AdaBoost and Soft voting classifier were analyzed and compared to the accuracy of all the algorithms. The comparative analysis showed that the Voting Classifier performed best in predicting most of the diseases, and Support Vector Machine and AdaBoost also performed well in terms of Precision, Sensitivity, and Specificity metrics. SVM achieved a classification accuracy of 91.9% for Alzheimer’s Disease, 88.7% for Lung Cancer, while the Voting Classifier, which is an ensemble of the mentioned ML algorithms, achieved accuracy of 77% for Heart Disease, 78.2% for Cardiovascular Disease, 91.4% for Breast Cancer, and 95% for Stroke, outperforming the other machine learning algorithms. The classification accuracy of the Voting Classifier was similar to that of SVM and AdaBoost in multi-disease prediction