PEMODELAN ARTIFICIAL NEURAL NETWORK PERAMALAN TINGKAT KENAIKAN JABATAN BERDASARKAN KINERJA PEGAWAI MENGGUNAKAN BACKPROPAGATION

Abstract

At PT JNE Pematangsiantar, this large capacity for delivery and service of goods and services clearly requires good human resources. Information about organizational performance is a very important thing used to evaluate whether the performance process carried out by the organization so far has been in line with the expected goals or not. In this study, a prediction system for the level of promotion based on performance will be designed by taking into account the factors of the work of the employee and the results of the assessment will be used to calculate the eligibility of an employee, intended to provide employee incentives or to retain the employee. An artificial neural network is an information processing system that is designed to imitate the workings of the human brain in solving a problem by carrying out the learning process through changes in the weight of its synapses. In this study the author uses the Backpropagation algorithm. Backpropagation gradient descent algorithm to minimize the square of the output error. There are three stages that must be carried out in network training, namely the forward propagation stage, the reverse propagation stage, and the weight and bias change stage. The results of this study indicate that the best architectural model (5-2-1) with MSE 0.004865154, epoch 126 with an accuracy rate of 70%. With the best prediction in predicting the promotion of employee positions at JNE.