|
A Study of Machine Learning for Crude Palm Oil Price Prediction |
|---|---|
| รหัสดีโอไอ | |
| Creator | Datchakorn Tancharoen |
| Title | A Study of Machine Learning for Crude Palm Oil Price Prediction |
| Contributor | Kanya Siripirom |
| Publisher | The Association of Council of IT Deans (CITT) |
| Publication Year | 2566 |
| Journal Title | Journal of Information Science and Technology |
| Journal Vol. | 13 |
| Journal No. | 2 |
| Page no. | 1-12 |
| Keyword | Baseline Model, LSTM, Classic Model, MAPE, Price Prediction |
| URL Website | https://tci-thaijo.org/index.php/JIST |
| Website title | Journal of Information Science and Technology |
| ISSN | 2651-1053 |
| Abstract | This paper focuses on the prediction of Crude Palm Oil prices in Thailand and find the best model for price prediction. Accurate price forecasting is essential for stakeholders in the palm oil industry to make informed decisions and manage risks effectively. In this comparative study between the Baseline Model, Recurrent Neural Network-Long Short-Term Memory, and the Classic models. The Classic model are Regression Model, Exponential Smoothing, Random Forest, LightGBM Model, and Theta Method. The testing will be conducted by comparing performance of models for the crude palm oil price prediction using time series and this is only one factor in this experiment. For the evaluation using the Mean Absolute Percentage Error (MAPE) metric to measure the percentage deviation between the projected and actual prices. It provides insights into the performance of the model. This research found that the baseline model has the lowest MAPE value is 0.08128. Therefore, the Baseline model provides the best accuracy for forecasting crude palm oil prices in Thailand. |
