Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal JURIKOM (Jurnal Riset Komputer)
Bahri, Cheisya Andini
Unknown Affiliation
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Perbandingan Kinerja LSTM, Random Forest, dan SVR Berbasis Knowledge Discovery untuk Prediksi Harga Beras Sumatera Selatan Bahri, Cheisya Andini; Tania, Ken Ditha
JURNAL RISET KOMPUTER (JURIKOM) Vol. 12 No. 5 (2025): Oktober 2025
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/jurikom.v12i5.9140

Abstract

Rice is a primary staple food in Indonesia, particularly in South Sumatra Province. In February 2024, BBC News Indonesia reported that the price of premium rice surged to Rp18,000 per kilogram, marking the highest price in the country’s history. To anticipate and predict similar spikes in the future, this study applies a Knowledge Discovery approach and compares three machine learning models: LSTM, Random Forest, and SVR. The approach follows the stages of data selection, cleaning, transformation, modeling, and evaluation to uncover hidden patterns in historical data. The dataset, obtained from the official PIHPS Nasional website, consists of 1,412 daily rice price records from January 2020 to May 2025. Model performance was evaluated using MAPE, MAE, and RMSE metrics. The findings indicate that the SVR model outperformed LSTM and Random Forest, delivering the most accurate results. For the Super Quality II rice category, SVR achieved a MAPE of 0.00 percent, MAE of 40.93, and RMSE of 52.54. SVR also consistently produced the lowest prediction errors in other categories, such as Low Quality I (MAE 59.39) and Medium Quality I (MAE 38.92). This research is expected to serve as a foundation for developing machine learning–based food price monitoring systems to support more responsive policies and maintain rice price stability in the future.
Co-Authors Ken Ditha Tania