Author: Mustakim, Agus Buono, and Irman Hermadi

Publish: Journal of Computer Science and Information

Abstract:

The largest region that produces oil palm in Indonesia has an important role in improving the welfare an economy of the society. Oil palm production has increased significantly in Riau Province in every period. To determine the production development for the next few years, we proposed a prediction of the production results. The dataset were taken to be the time series data of the last 8 years (2005-2013) with the function and benefits of oil palm as the parameters. The study was undertaken by comparing the performance of Support Vector Regression (SVR) method and Artificial Neural Network (ANN). From the experiment, SVR resulted the better model compared to the ANN. This is shown by the correlation coefficient of 95% and 6% for MSE in the kernel Radial Basis Function (RBF), whereas ANN resulted only 74% for R2 and 9% for MSE on the 8th experiment with hidden neuron 20 and learning rate 0,1. SVR model generated predictions for next 3 years which rose 3%-6% from the actual data and RBF model predictions.

Conclusion:

From the conducted research, it can be concludedthat model SVR is better than model ANNfor oil palm production prediction’s case in Riau.Model ANN got the best value of determination coefficient (R2) 74% with galat error 9% on the 8th experiment, while SVR on the RBF kernel produ-ced a smaller galat is 6% and also R2 is bigger than ANN that produced 95%. A very huge difference of determination coefficient value proved that by using time series data, model SVR is more superior compared to model ANN. Prediction results for next three years gradually in normal form as many as 3%-6%. Prediction results do not reckon the nature or other factors in the field that could effect production in each period.

Sumber Gambar