Prediksi Harga Crude Palm Oil (CPO) Dengan Model Transformers

Hanifah, Khairunnisa Isma (2024) Prediksi Harga Crude Palm Oil (CPO) Dengan Model Transformers. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06111740000054-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 July 2026. Download (2MB) | Request a copy

Abstract

Prediksi harga Crude Palm Oil (CPO) memiliki peranan krusial dalam industri perkebunan kelapa sawit dan perdagangan komoditas. Penelitian ini bertujuan untuk menerapkan model Transformers dalam memprediksi harga CPO dan mengetahui akurasinya. Model Transformers adalah arsitektur jaringan saraf tiruan yang telah sukses digunakan dalam berbagai aplikasi Natural Language Processing (NLP) dan analisis time series. Hasil penelitian menunjukkan bahwa model Transformers memberikan hasil yang objektif dalam memprediksi harga CPO dalam jangka waktu tertentu. Akurasi model diukur menggunakan berbagai evaluasi performa model seperti Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Squared Error (MSE), dan Mean Absolute Percentage Error (MAPE) menunjukkan nilai yang rendah, yaitu MSE sebesar 0.00019, MAE sebesar 0.00993, MAPE sebesar 2.716%, dan RMSE sebesar 0.01363. Hasil ini mengindikasikan bahwa model Transformers memberikan prediksi harga CPO yang mendekati nilai aktual
===================================================================================================================================
Crude Palm Oil (CPO) price predictions have a crucial role in the palm oil plantation industry and commodity trading. This research aims to apply the Transformers model to predict CPO prices and determine its accuracy. The Transformers model is an artificial neural network architecture that has been successfully used in various Natural Language Processing (NLP) applications and time series analysis. The research results show that the Transformers model provides objective results in predicting CPO prices within a certain time period. Model accuracy is measured using various model performance evaluations such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Squared Error (MSE), and Mean Absolute Percentage Error (MAPE) showing low values, namely MSE of 0.00019, MAE of 0.00993, MAPE of 2.716%, and RMSE of 0.01363. These results indicate that the Transformers model provides predictions of CPO prices that are close to actual values

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Prediksi, Crude Palm Oil (CPO), Model Transformers; Prediction, Crude Palm Oil (CPO), Transformers Model
Subjects:	Q Science > Q Science (General) Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA280 Box-Jenkins forecasting Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	Khairunnisa Isma Hanifah
Date Deposited:	19 Feb 2024 08:40
Last Modified:	19 Feb 2024 08:40
URI:	http://repository.its.ac.id/id/eprint/107527

Actions (login required)

View Item