Classification of Family Welfare Card Recipients in Makassar City Using Decision Tree Algorithms

Abstract

This study aims to analyze the factors influencing the determination of recipients of the Family Welfare Card (KKS) program in Makassar City and evaluate the level of accuracy of the decision tree model in the classification process. The KKS program is a government effort to accelerate poverty alleviation, so it is important to ensure that the selection process for program recipients is carried out on target. The decision tree method is used in this study because of its ability to simplify the decision-making process through an easy-to-understand tree structure. This study utilizes KKS recipient data with various variables, such as income, number of dependents, employment status, asset ownership, and education level, to build a classification model. The results of the study indicate that the variable of the Head of Household's (KRT) Highest Education Level (X4) has the highest level of importance in determining KKS recipients, followed by the variable Number of Family Members (X1), and the variable Ownership of Residential Buildings (X5). The decision tree model that was built has an accuracy level of 84.21%, which states the model's ability to classify KKS recipients effectively. This study also provides insight into the description of factors influencing KKS receipts, which can be used as a basis for formulating more efficient and targeted policies.