OPTIMASI ALGORITMA C4.5 DENGAN MENGGUNAKAN ALGORITMA GENETIKA UNTUK MEMPREDIKSI MEREK METERAN AIR PAM

Abstract

The Drinking Water Company has endeavored to provide services to meet the need for clean water, but in its journey often received complaints from the community or pam customers. Customer complaints can be caused from damage to the meter Water causes water to die into the house, a leakage in the meter count, leakage in the meter body or excess usage when checking takes place in each house, the number of meters damaged causes the stock in the warehouse to be incompatible with the usage needs new installation and use due to meter damage. Due to the lack of clean water in the area, there is a demand from the community to optimize their needs, but the fact is that it cannot be optimized yet because the water pipeline has not yet been installed. Extracting large amounts of data is usually called data mining. Because there is no research on meter brands using the C4.5 Algorithm method and is optimized using the Genetic Algorithm. In conducting this test the tools used are Rapidminer. The results obtained using the C4.5 algorithm without optimization are 63.33% and the results obtained by the C4.5 algorithm and genetic algorithm optimization are 90.00% or an increase of 26.67% from the C4.5 algorithm without optimization.

Keywords: Data Mining, Meter Brand, C4.5 Algorithm, Genetic Algorithm.