Analisis Komparatif Model Regresi Linier dan Polinomial pada Small Dataset untuk Prediksi Timbulan Sampah
Indonesia
Keywords:
Small Dataset, Waste Prediction, Linear Regression, Polynomial RegressionAbstract
Bandung Regency faces severe waste management challenges, generating an average daily volume of 1,300 tons without possessing an independent landfill facility. Consequently, accurate data-driven prediction is crucial for strategic infrastructure planning. However, the application of machine learning algorithms at the regional level is often significantly constrained by the scarcity of historical data (small datasets), which introduces high risks of overfitting and statistical bias. This study aims to evaluate prediction modeling strategies using a limited annual dataset (n=4) spanning from 2021 to 2024, sourced from Open Data Jabar. The methodology compares Linear Regression and Second-Degree Polynomial Regression algorithms by employing a full training approach combined with descriptive validation based on the Rate of Change (RoC) analysis. The results indicate that while Polynomial Regression achieves superior statistical performance with an R-squared of 0.9958 and an RMSE of 3,144.59 tons, trend analysis reveals clear signs of overfitting, as the model predicts an implausible deceleration in future waste growth that contradicts demographic realities. Conversely, Linear Regression (R2 = 0.9756) provides more stable and consistent trend estimates. Therefore, this study recommends Linear Regression as a more robust model for policy planning under limited data conditions, projecting waste generation to reach 467,964.01 tons in 2025, serving as an early warning for urgent capacity expansion.
References
“SIPSN - Sistem Informasi Pengelolaan Sampah Nasional.”
F. M. Aprileni and A. Nurhayati, “Penanganan Sampah Rumah Tangga Di Desa X Kabupaten Bandung,” J. Ris. Kesehat. Poltekkes Depkes Bandung, vol. 16, no. 2, pp. 702–707, 2024, doi: 10.34011/juriskesbdg.v16i2.2635.
N. Alhathlaul et al., “Assessing Waste Management Using Machine Learning Forecasting for Sustainable Development Goal Driven,” vol. 6, pp. 1–21, 2025.
L. Choi, Wonjun; Sangwon, “Performance evaluation of deep learning architectures for load and temperature forecasting under dataset size constraints and seasonality,” Energy Build., p. 288, 2023.
S. A. L. Satriyo, Adi Rizky Pratama, and Rahmat, “Perbandingan metode linear regresi dan polynomial regresi untuk memprediksi harga saham studi kasus Bank BCA,” INFOTECH J. Inform. Teknol., vol. 4, no. 1, pp. 59–70, 2023, doi: 10.37373/infotech.v4i1.602.
J. K. Kang, Y. J. Lee, C. Y. Son, S. J. Park, and C. G. Lee, “Alternative assessment of machine learning to polynomial regression in response surface methodology for predicting decolorization efficiency in textile wastewater treatment,” Chemosphere, vol. 370, no. December 2024, p. 143996, 2025, doi: 10.1016/j.chemosphere.2024.143996.
N. M. Farhan and B. Setiaji, “Perbandingan Algoritma Regresi Linear dengan Algoritma Backpropagation Estimasi Timbulan Sampah di Sulawesi Utara,” Indones. J. Comput. Sci., vol. 12, no. 2, pp. 284–301, 2023, [Online]. Available: http://ijcs.stmikindonesia.ac.id/ijcs/index.php/ijcs/article/view/3135
S. S. M. A. B. Bagus Irawan1*, “Review penggunaan machine learning dalam optimalisasi pengelolaan sampah perkotaan: studi literatur terkini,” Pros. Semin. Nas. Sains dan Teknol. Seri 02, vol. 1, no. 2, pp. 543–553, 2024.
A. Widianti and I. Pratama, “Penanganan Missing Values Dan Prediksi Data Timbunan Sampah Berbasis Machine Learning,” Rabit J. Teknol. dan Sist. Inf. Univrab, vol. 9, no. 2, pp. 242–251, 2024, doi: 10.36341/rabit.v9i2.4789.
G. G. Montgomery, Douglas C., Peck, Elizabeth A., & Vining, Introduction to Linear Regression Analysis (6th ed.). Wiley, 2021.
N. R. S. Jayadi, Puguh; Hidayati, “Data-Driven Approach of Machine Learning Techniques for Forecasting on Small Dataset: A Case Study of City Waste,” IEEE, 2025.
U. Rohman, J. Sunupurwa Asri, H. D. Ariessanti, and S. Wahyu, “Perbandingan Algoritma Regresi Linear Sederhana dan Regresi Polinomial Dalam Prediksi Jumlah Penumpang Kereta Api Di Wilayah Jabodetabek,” ProTekInfo(Pengembangan Ris. dan Obs. Tek. Inform., vol. 12, no. 1, pp. 7–15, 2025, doi: 10.30656/protekinfo.v12i1.9842.
R. O. Budianto and L. P. Ghanistyana, “Peran Komunikasi Politik dalam Kampanye Isu Lingkungan: Studi Kasus pada Kebijakan Pengelolaan Sampah di Indonesia,” J. Bisnis dan Komun. Digit., vol. 2, no. 1, p. 11, 2024, doi: 10.47134/jbkd.v2i1.3219.
Rahmayani, R., Firmansyah, E., & Hikmah, H. U. (2025). Inovasi Layanan Antar Jemput Paket Surat PT Pos Indonesia Berdasarkan Penjualan dan Minat Beli. JOVISHE: Journal of Visionary Sharia Economy, 4(1), 33-47.
Firmansyah, E., Rahman, A. B. A., & Subiyakto, A. A. (2023). Pengukuran Kesiapan Kota Cerdas Berdasarkan SNI ISO 37122: 2019. Infoman's: Jurnal Ilmu-ilmu Informatika dan Manajemen, 17(2).
Zulfikar, W. B., Irfan, M., Ghufron, M., Jumadi, J., & Firmansyah, E. (2020). Marketplace affiliates potential analysis using cosine similarity and vision-based page segmentation. Bulletin of Electrical Engineering and Informatics, 9(6), 2492-2498.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Infoman's : Jurnal Ilmu-ilmu Informatika dan Manajemen
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License.
