PREDIKSI PENDAPATAN ASLI DAERAH (PAD) KABUPATEN LANGKAT MENGGUNAKAN METODE BACKPROPAGATION NEURAL NETWORK

Abstract

The District Government of Langkat in this case regulates and manages Regional Original Income (PAD), for example, taxes can be increased by intensification and extentification. Extensification is an expansion of the type of tax, but several studies show that controlling the potential by expanding the type of tax does not stimulate interest and even creates reluctance for investors to invest in the area. Intensification is an effort to increase tax collection. This effort requires the ability of the regions to be able to correctly identify local revenue and the ability to collect taxes based on benefits and principles of justice. By inputting training data and training targets, the artificial neural network to predict the number of Langkat Regency using the backpropagation method can predict the amount of PAD in Langkat Regency. The artificial neural network system can recognize training data and target data with an iteration of 488 target error of 0.5 and a leraning rate of 0.1, resulting in a prediction of the amount of PAD in 2020, which is Rp. 140,948,000,000.