Identification of Acute Respiratory Infections in Toddlers Based on the Chi-Square And Naive Bayes Methods

Acute Respiratory Infections usually attack the respiratory tract of toddlers, both the upper and lower respiratory tracts, because the body's defense system against viruses that cause infection has not yet been formed. And usually, parents will know if the baby's condition is very chronic so that the baby experiences complications. This causes the need for a system that can assist in the early detection of respiratory tract infections. This study proposes the Chi-Square and Naive Bayes (NB) method. The Chi-Square method is a feature selection method to reduce features that have no effect. At the same time, the NB method is a prediction method that performs a simple probability-based identification process based on the application of the Bayes theorem with the assumption of strong independence. The contribution of this study is to determine respiratory tract disease in infants using the chi-square feature selection and the NB method, which can assist parents in detecting respiratory tract infections. From the tests that have been carried out using 120 datasets with 90 as training data and 30 as test data, the accuracy is 75.833%. This proves that the Chi-Square and NB methods are able to identify respiratory tract infections.