A Comparative Analysis of Machine Learning Models in Predicting Blood Donation Behavior

The prediction of blood donation behavior is essential for improving donor recruitment and retention strategies within healthcare systems. This study performed a comparative analysis of three machine learning models such as Logistic Regression, Random Forest and Support Vector Machine (SVM) to predict blood donation behavior based on blood donation history data. The primary goal was to conduct a comparative analysis of three machine learning models. The study employed a comprehensive dataset that included various features related to donation history of potential donors. The models were evaluated using several key performance metrics, including accuracy, precision, recall, F1 score, and ROC-AUC, which provide assessing their predictive capabilities. The findings of the analysis indicated that the Random Forest model significantly outperformed the other two algorithms, achieving an accuracy of 92% and a ROC-AUC score of 0.93. This superior performance was attributed to Random Forest’s ability to capture complex interactions within the dataset, making it particularly effective for this type of predictive modeling. In contrast, SVM and Logistic Regression demonstrated lower accuracy and predictive power, highlighting their limitations in this context. The results of this study highlight the potential of machine learning techniques to improve blood donation strategies. By utilizing advanced predictive modeling, healthcare organizations can refine their outreach efforts, ultimately increasing donation rates and addressing critical public health needs. This research contributes to the expanding field of predictive analytics in healthcare, providing valuable insights that can inform future initiatives aimed at improving blood donation behaviors.