Machine learning algorithms for important feature evaluation and prediction of severe hand-foot-mouth disease in hunan province, China

Hand, foot, and mouth disease(HFMD) is an infectious disease of the intestines that damages people's health, severe cases could lead to cardiorespiratory failure or death. Therefore, the evaluation of important features and prediction for severe HFMD is critical for early prevention and control of the disease. With this goal in mind, 658,689 cases which include 6,579 severe cases were assessed. In this research-in-progress paper, we are trying to establish an easy, automatic and efficient server HFMD prediction system based on hospital case data and meteorological data, and Random Forests and Adaboost algorithm were utilized in this paper for feature importance evaluation. Preliminary experimental result shows that our model can evaluate the importance of features but parameters still need further adjustment for predictions of severe HFMD.