The free electricity installation program initiated by the ESDM Office of South Kalimantan Province requires accurate eligibility criteria so that assistance is right on target. One of the main challenges is the lack of a method that considers the weight of interests in determining recipients. This study applies the Naive Bayes method, a probability-based statistical technique, to predict the eligibility of aid recipients. The recipient data was divided 60:40 for model training and testing. The evaluation results using a confusion matrix showed an accuracy of 93.94%, higher than the weighting-based method which only reached 86.59%. These findings prove that Naive Bayes is effective in increasing the accuracy and reliability of aid acceptance decision making. Thus, this method is worth considering in similar programs to ensure efficiency and targeting accuracy, as well as providing a stronger foundation for the implementation of future social assistance programs.Keywords: Naive Bayes; Free Electricity; Probability AbstrakProgram pemasangan listrik gratis yang diinisiasi Dinas ESDM Provinsi Kalimantan Selatan membutuhkan kriteria kelayakan yang akurat agar bantuan tepat sasaran. Salah satu tantangan utama adalah kurangnya metode yang mempertimbangkan bobot kepentingan dalam menentukan penerima. Penelitian ini menerapkan metode Naive Bayes, sebuah teknik statistik berbasis probabilitas, untuk memprediksi kelayakan penerima bantuan. Data penerima bantuan dibagi 60:40 untuk pelatihan dan pengujian model. Hasil evaluasi menggunakan confusion matrix menunjukkan akurasi 93,94%, lebih tinggi dibandingkan metode berbasis pembobotan yang hanya mencapai 86,59%. Temuan ini membuktikan bahwa Naive Bayes efektif meningkatkan akurasi dan keandalan dalam pengambilan keputusan penerimaan bantuan. Dengan demikian, metode ini layak dipertimbangkan dalam program serupa untuk memastikan efisiensi dan ketepatan sasaran, serta memberikan dasar yang lebih kuat bagi pelaksanaan program bantuan sosial di masa depan.
Copyrights © 2024