In recent years, the use of machine learning algorithms into public health procedures has demonstrated great promise in forecasting the emergence of communicable illnesses. This study compares machine learning methods, such as Random Forest, Logistic Regression, Support Vector Machine (SVM), and k-Nearest Neighbors (k-NN), for improving communicable disease surveillance using data from the Province of Marinduque. The information ranges from 2015 to 2019 and includes illnesses such as Hand Foot Mouth Disease, Dengue, Typhoid, Influenza, Chikungunya, Rabies, Measles, Meningitis, Hepatitis, and Acute Bloody Diarrhea. The monthly morbidity rate acts as the criteria variable for prediction. The study used this dataset to establish the best effective machine learning model for illness prediction by measuring its accuracy, precision, recall, and F1-score. The Random Forest model had the best accuracy of the models tested, demonstrating its promise for reliable illness incidence prediction. To evaluate the models' performance, the study employs the Evaluation Metrics of Accuracy, Precision, Recall, and F1-score.

The findings of this study can help to improve communicable disease monitoring systems, eventually assisting in the prompt detection and control of disease outbreaks.