Enhancing Monkeypox Detection: A Machine Learning Approach to Symptom Analysis and Disease Prediction

Monkeypox disease, caused by the monkeypox virus, is a highly communicable disease and has prompted the World Health Organization to declare it as a Public Health Emergency. Early recognition of the disease is imperative to stop the community transmission of monkeypox. The objective of the study is to detect monkeypox based solely on its clinical symptoms consisting of boolean and categorical features through the use of various machine learning models: decision tree, logistic regression, Naïve Bayes, support vector machine, random forest, and Adaptive Boosting. The best performing model was obtained by the support vector machine with the highest area under the precision recall curve of 79.67%, 88.35% recall, 72.86% precision and 84.53% F2-score under the baseline model configuration (without feature selection, without class imbalance correction). The superior recall of the best model highlights its ability to correctly diagnose monkeypox resulting in a much lower false negatives or fewer missed cases of monkeypox crucial in halting its community spread. The integration of these models in the clinical setting can serve as a decision support tool for prompt monkeypox recognition. As such, swift medical intervention to those afflicted can be administered as well as prompt institution of quarantine measures to prevent the community transmission of monkeypox.