Prediksi Realisasi PBJT Menggunakan KNN Regression
DOI:
https://doi.org/10.54367/kakifikom.v8i2.6985Keywords:
PBJT, KNN Regression, forecasting, tax revenue, data miningAbstract
Purpose: This study predicts the realization of Certain Goods and Services Tax (PBJT) revenue in the food and beverage sector at the Regional Financial and Revenue Management Agency of Simalungun Regency using K-Nearest Neighbor Regression. Design/methods/approach: The study applies a quantitative data mining approach using monthly historical data from 2021 to 2025. Input variables consist of year, month, and number of tax objects, while monthly PBJT realization is used as the output variable. The modeling stages include dataset construction, Min-Max Scaling, time-based data splitting, K-value validation, KNN Regression training, and model evaluation using MAE, MSE, RMSE, MAPE, and estimation accuracy. Findings/results: Validation on 2024 data selected K=1 as the best parameter with a MAPE of 40.64%. Final testing on 2025 data produced MAE of IDR 180,533,900, RMSE of IDR 205,995,400, MAPE of 51.40%, and estimation accuracy of 48.60%. Conclusions: KNN Regression can be implemented as an initial forecasting tool, but its prediction error remains high; therefore, additional fiscal and economic variables are needed before the model is used for target-setting decisions.References
K. Ananda Mustari, P. Assiroj, B. Hartati, and F. Samuel, "Implementasi Data Mining pada Instansi Pemerintahan (Systematic Literature Review)," Jurnal Mahasiswa Teknik Informatika, vol. 8, no. 3, 2024.
V. Pamisetty, L. Pandiri, S. Singireddy, V. N. Annapareddy, and H. K. Sriram, "Leveraging AI, Machine Learning, and Big Data for Enhancing Tax Compliance, Fraud Detection, and Predictive Analytics in Government Financial Management," Migration Letters, S5, pp. 1770-1784, 2022.
M. Suyal and P. Goyal, "A Review on Analysis of K-Nearest Neighbor Classification Machine Learning Algorithms Based on Supervised Learning," International Journal of Engineering Trends and Technology, vol. 70, no. 7, pp. 43-48, 2022, doi: 10.14445/22315381/IJETT-V70I7P205.
W. Wahdanianti, N. Nasrullah, and A. T. Syahriani, "Strategi Optimalisasi Penerimaan Pajak Restoran dalam Peningkatan Pendapatan Asli Daerah (PAD) di BAPENDA Kabupaten Sinjai," RIGGS: Journal of Artificial Intelligence and Digital Business, vol. 5, no. 1, pp. 13254-13263, 2026, doi: 10.31004/riggs.v5i1.7822.
K. N. Kalangi, S. Pangerapan, and I. G. Suwetja, "Analisis Penerapan UU Nomor 1 Tahun 2022 tentang Hubungan Keuangan antara Pemerintah Pusat dan Pemerintahan Daerah dalam Tata Cara Pemungutan PBJT," Riset Akuntansi dan Portofolio Investasi, vol. 3, no. 2, pp. 466-475, 2025, doi: 10.58784/rapi.369.
M. Marhamah and I. Harsono, "Analisis Pendapatan Asli Daerah untuk Mengukur Kemandirian Keuangan Daerah di Kota Mataram Tahun 2018-2023," Jurnal Syntax Imperatif, vol. 6, no. 3, pp. 349-360, 2025, doi: 10.54543/syntaximperatif.v6i3.699.
R. Wirth and J. Hipp, "CRISP-DM: Towards a Standard Process Model for Data Mining," in Proc. 4th Int. Conf. Practical Applications of Knowledge Discovery and Data Mining, 2000.
E. Bisong, "Google Colaboratory," in Building Machine Learning and Deep Learning Models on Google Cloud Platform. Berkeley, CA: Apress, 2019.
S. Wiyono and T. Abidin, "Implementation of K-Nearest Neighbour (KNN) Algorithm to Predict Student's Performance," Jurnal Simetris, vol. 9, no. 2, 2018.
Republik Indonesia, Undang-Undang Nomor 1 Tahun 2022 tentang Hubungan Keuangan antara Pemerintah Pusat dan Pemerintahan Daerah, 2022.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 KAKIFIKOM (Kumpulan Artikel Karya Ilmiah Fakultas Ilmu Komputer)
This work is licensed under a Creative Commons Attribution 4.0 International License.
