Perbandingan Model XGBoost, LSTM, Dan Neural Prophet Untuk Prediksi Harga Cabai Rawit Merah Di Jawa Timur

Anamsyah, Hafid Alfa (2026) Perbandingan Model XGBoost, LSTM, Dan Neural Prophet Untuk Prediksi Harga Cabai Rawit Merah Di Jawa Timur. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (cover) 20081010189.-cover.pdf Download (690kB)
	Text (bab 1) 20081010189.-bab1.pdf Download (173kB)
	Text (bab 2) 20081010189.-bab2.pdf Restricted to Repository staff only until 26 June 2028. Download (166kB)
	Text (bab 3) 20081010189.-bab3.pdf Restricted to Repository staff only until 26 June 2028. Download (570kB)
	Text (bab 4) 20081010189.-bab4.pdf Restricted to Repository staff only until 26 June 2028. Download (1MB)
	Text (bab 5) 20081010189.-bab5.pdf Download (146kB)
	Text (daftar pustaka) 20081010189.-daftarpustaka.pdf Download (151kB)

Abstract

The price of red cayenne pepper in Indonesia, especially in East Java Province, often experiences significant fluctuations that have an impact on food inflation and the economic stability of the community. Therefore, accurate price predictions are needed to help decision-making in the management of the food supply chain. This study aims to compare the performance of three machine learning and deep learning models, namely XGBoost, LSTM, and Neural Prophet, in predicting the price of red cayenne pepper based on historical data from September 2024 to August 2025 obtained from the official website of the Siskaperbapo of the East Java Province Department of Industry and Trade (DISPERINDAG). The analysis was carried out using a quantitative approach using time series forecasting data, with performance evaluation through RMSE, MAE, and MAPE metrics. The results showed that the XGBoost model provided the best accuracy with RMSE values of 0.52, MAE 0.40, and MAPE of 1.42%, compared to XGBoost and Neural Prophet. Based on the overall results of the analysis, it can be concluded that the model selection, the length of the data, and the proportion of data sharing have a significant effect on the accuracy of price predictions, where XGBoost is the best model because it produces the most accurate and stable predictions, and the use of longer datasets and larger training data tends to improve model performance. This research contributes to the development of artificial intelligence-based food price prediction as a basis for price stabilization policy decision-making in the agricultural sector.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Diyasa, I Gede Susrama Mas	NIDN0019067008	igusrama.if@upnjatim.ac.id
Thesis advisor	Sihananto, Andreas Nugroho	NIDN0012049005	andreas.nugroho.jarkom@upnjatim.ac.id

Subjects:

T Technology > T Technology (General)

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Hafid Alfa Anamsyah

Date Deposited:

26 Jun 2026 06:49

Last Modified:

26 Jun 2026 07:09

URI:

https://repository.upnjatim.ac.id/id/eprint/54201