Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Jurnal Riset dan Aplikasi Mahasiswa Informatika (JRAMI)
Valencia Patrice Gracia Pangaribuan
Universitas Esa Unggul

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

Found 1 Documents
Search

 1 
Prediksi Harga Komoditas Hortikultura Pangan Strategis Menggunakan Random Forest dengan Analisis SHAP Valencia Patrice Gracia Pangaribuan; Bayu Sulistiyanto Ipung Sutejo; Gerry Firmansyah; Arief Ichwani
Jurnal Riset dan Aplikasi Mahasiswa Informatika (JRAMI) Vol. 7 No. 03 (2026): Jurnal Riset dan Aplikasi Mahasiswa Informatika (JRAMI)
Publisher : Program Studi Teknik Informatika, FTIK, Universitas Indraprasta PGRI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30998/jrami.v7i03.1804

Abstract

Price fluctuations in strategic horticultural food commodities can affect food and economic stability. This study aims to predict the prices of shallots, garlic, red chilies, and bird’s eye chilies in Central Java Province using Random Forest Regression based on historical prices and climate variables, and to interpret climate variable contributions using Shapley Additive Explanations (SHAP). Price data were obtained from the National Strategic Food Price Information Center (PIHPS), while climate data were collected from four BMKG stations during 2018–2025. The model used 16 features, including historical prices, average temperature (TAVG), average humidity (RH_AVG), rainfall (RR), rolling features, and calendar variables. Evaluation was conducted using MAE, RMSE, and MAPE at H+1, H+7, and H+30 forecasting horizons. The results show that the best performance was achieved for garlic price prediction at H+1, with a MAPE of 1.69%, while the highest error occurred for red chili at H+30, with a MAPE of 26.72%. SHAP analysis shows that TAVG was dominant in chili commodities, while RR and RH_AVG had greater influence on shallots and garlic. These findings indicate that Random Forest and SHAP can support short-term price prediction with interpretable climate variable contributions.
Co-Authors Bayu Sulistiyanto Ipung Sutejo Gerry Firmansyah Ichwani, Arief