Komparasi Algoritma C4.5 Dan C4.5 Berbasis Pso Untuk Prediksi Jumlah Penggunaan BBM Perbulan Pada Kantor Dinas Lingkungan Hidup Dan Kebersihan Kabupaten Lombok Timur

Abstract

East Lombok Regency is one of the second level regions in West Nusa Tenggara Province which is located on the east side of Lombok Island. The capital city of East Lombok Regency is the city of Selong, where all government agencies are based in this city. One of them is the Department of Environment and Hygiene of East Lombok Regency. In carrying out operational duties at the Office of Environment and Hygiene the operational vehicle requires that the fuel oil is a subsidy from the government. Therefore, the use of BBM every day must be recorded properly so that it can be predicted the amount of fuel usage every month. However, the Office of the Environment and Hygiene Office has difficulty in processing such data in large quantities. Predicted information on fuel use is needed by the head of the agency to assist in making decisions or policies. Of these problems the right data mining technique to use is classification. One method of classification of data mining is the decition tree algorithm (C4.5) or called the decision tree. The decition tree (C4.5) algorithm has weaknesses in reading large amounts of data, so researchers use weighting by applying Particle Swarm Optimization (PSO) for attribute selection to increase the accuracy of C4.5. Thus the researcher will utilize data mining software in applying a comparison of the decition tree (C4.5) and C4.5 algorithms based on Particle Swarm Optimization (PSO) to get the best accuracy value in predicting the amount of monthly use of fuel oil at the Service Office Environment and Cleanliness of East Lombok Regency. DOI : 10.29408/jit.v2i1.1173