Prediksi Penggunaan Obat Peserta Jaminan Kesehatan Nasional Menggunakan Algoritma Naïve Bayes Classifier

Abstract

Currently, most of the patients seeking treatment at the hospital use the National Health Insurance (JKN) organized by the Healthcare and Social Security Agency (BPJS Kesehatan). In some hospitals, the figure is above 80%. Considering the very high number of BPJS Kesehatan participant seeking treatment at the hospital, a good data management method is needed, especially regarding the management of drug. Drug supply needs to be analyzed from time to time so that it can help predict future needs. An adequate supply of drugs and as needed is one of the things that affect service to patients. The availability of sufficient stock is expected to accelerate service to patients so that they do not have to wait long. Patients who are served quickly are expected to be satisfied. The impact of this patient satisfaction will increase the number of patient visits to the hospital. To support this, it is necessary to create a system that can estimate drug needs. The system can predict drug demand by using drug sales data to JKN participant patients for five years. Drug data used as research samples and then processed using an algorithm is the Naive Bayes Classifier. The Naive Bayes Classifier method is a method used to predict future opportunities using the basis of previous experience. A distinctive feature of this method is that it uses a very strong assumption of the independence of each event. While software testing uses the User Acceptance Test (UAT) model. Based on testing using this method, the system can be well received by users with a score of 78.64% (good).