Application of the C5.0 Algorithm to Determine the Eligibility of BPJS Contribution Assistance Recipients in the National Health Insurance Program

The National Health Insurance (JKN) is a government program to provide health insurance to all Indonesian citizens. In its implementation, determining the eligibility of premium assistance recipients still faces challenges regarding accuracy and efficiency. This study aims to implement the C5.0 algorithm in classifying the eligibility of JKN premium assistance recipients and measure the accuracy of the resulting classification model. The method used is the C5.0 algorithm with the boosting technique with 10 trials and global pruning, with a 70:30 training and testing data split ratio, minimum cases of 2, and confidence level of 0.2. The research results show that the classification model performs excellently with an accuracy rate of 94.27%, precision of 83.78%, Recall of 90.29%, F1-Score of 86.92%, and AUC of 92.81%. The Specificity, reaching 95.34%, demonstrates the model's reliability in identifying ineligible participants for assistance. With a total algorithm execution time of 0.2224 seconds, these results indicate that the C5.0 algorithm can be implemented as an effective decision support system in determining the eligibility of JKN premium assistance recipients.