PENERAPAN DATA MINING DALAM PREDIKSI PENGANGGURAN BERDASARKAN TINGKAT PENDIDIKAN MENGGUNAKAN ALGORITMA BACKPROPAGATION NEURAL NETWORK
No Thumbnail Available
Date
2023-11
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Elfitra
Abstract
Unemployment is still widely found in Riau province. There are many factors that
influence unemployment, one of which is the level of education completed. Therefore,
unemployment prediction is needed to overcome the problem. Prediction is done by
using backpropagation neural network method. The purpose of this research is to
determine the best model in predicting unemployment based on education in Riau. The
best model is determined from the evaluation results of the smallest mean squared error
(MSE) and mean absolute percentage error (MAPE) values. From the test results, the
best architecture model is obtained, namely model C with an architecture of 8 input
layers, 4 hidden neurons, 1 output layer (8-4-1) with 0.01 learning rate. From the
evaluation result, model C produces MSE value of 0.0027 and MAPE value of 16.99%.
Thus, model C is able to predict unemployment based on education level in Riau well
and can be used for further prediction of unemployment based on education.
Description
Keywords
Backpropagation Neural Network, MAPE, MSE, Unemployment, Prediction
Citation
Perpustakaan