Non-Cash Food Assistance (BPNT) is social food assistance in the form of non-cash from the government which is given to Beneficiary Families (KPM) every month through an electronic account mechanism which is used only to buy food at traders or e-warongs. One of the difficulties that the government sometimes faces in distributing BPNT is that the distribution process is uneven and not on target. Therefore, it is necessary to carry out further analysis using a mathematical approach, so that we can determine the feasibility of a BPNT recipient prediction problem. Through the results of the data collection analysis, it can be seen whether residents are eligible to receive BPNT or not. Based on existing problems, a classification method is used to predict the eligibility of BPNT beneficiaries using two methods, namely the ID3 algorithm and the C4.5 algorithm. The ID3 algorithm produces an accuracy value of 90%, precision of 100%, and recall of 83.33%. The C4.5 algorithm produces an accuracy value of 80%, precision of 100%, and recall of 80%. The AUC/ROC value of the ID3 algorithm is 0.500, the classification is diagnosed in the AUC/ROC curve as failure or failure in classification. The C4.5 algorithm has an AUC/ROC value of 0.800, meaning that the classification is included in good classification. In this way, it can be concluded that the C4.5 algorithm has better results compared to the ID3 algorithm