Despite of the efforts of the Department of Health in the Philippines against the fight of the Acute Respiratory Infection or Pneumonia, this still becomes the top issues as its continually prompt to evolve the cases. This study aims to address the challenges of pneumonia using data mining in support of the Apriori Algorithm and classification-based association (CBA) model. To uncover patterns of pneumonia the study refined the 260 association rules out of 650,838 records that refined 10 associated clusters. With the support of 90:10 training set, the model commends a result of 93% for the True Positive Rate (TPR) and 7% for the False Positive Rate (FPR). The results of the TPR were optimized from 60%, 70%, and 83% in training and testing data. In this case, it narrowed down the variables from 42 to 26 that includes age, sex, immunization history, nutrition status, senior house living, hospital stays, family used of tobacco, and the presence of identified type of comorbidities. The findings provide a promising model for effective pneumonia patterns, paving the way for targeted support and interventions.