PENGEMBANGAN MODEL ALGORITMA RANDOM FOREST, EXTREME GRADIENT BOOST, DAN ARTIFICIAL NEURAL NETWORK UNTUK PENILAIAN TANAH MASSAL BERBASIS BIDANG MENGGUNAKAN HEDONIC PRICING MODELLING DI KOTA YOGYAKARTA

REZA ARISKA, 22314212 (2026) PENGEMBANGAN MODEL ALGORITMA RANDOM FOREST, EXTREME GRADIENT BOOST, DAN ARTIFICIAL NEURAL NETWORK UNTUK PENILAIAN TANAH MASSAL BERBASIS BIDANG MENGGUNAKAN HEDONIC PRICING MODELLING DI KOTA YOGYAKARTA. Diploma thesis, Politeknik Agraria STPN.

[img] Text
Reza Ariska_22314212.pdf
Download (1MB)

Abstract

Mass appraisal of land parcels is essential for supporting data-driven land policies, regional taxation, and spatial management. However, the conventional Hedonic Pricing Modelling (HPM) approach has limitations in capturing complex and nonlinear relationships among land value determinants. This study develops and compares the performance of Random Forest, Extreme Gradient Boosting (XGBoost), and Artificial Neural Network (ANN) models for parcel-based land value estimation using the HPM approach in Yogyakarta City. A quantitative approach was applied using 172 land parcels with variables representing physical characteristics, accessibility, location, environmental factors, and spatial planning attributes. The HPM model identified land area and frontage width as dominant factors, with an explanatory capability of R2 = 0.3731. Machine learning models improved prediction performance, with Random Forest achieving R2 = 0.6858 and MAE = 10.76, ANN achieving R2 = 0.5789 and MAE = 16.25, and XGBoost providing the best performance with R2 = 0.7487 and MAE = 9.85. XGBoost was identified as the most suitable model due to its ability to capture nonlinear patterns, provide stable predictions, and support parcel-based mass land valuation. Keywords: Mass Land Valuation, Hedonic Pricing Modelling, XGBoost, Random Forest, Artificial Neural Network.

Item Type: Thesis (Diploma)
Subjects: H Social Sciences > HD Industries. Land use. Labor > HD29 Pertanahan
Divisions: Prodi Diploma IV Pertanahan
Depositing User: yosep ka perpus
Date Deposited: 06 Oct 2026 03:43
Last Modified: 06 Oct 2026 03:43
URI: http://repository.stpn.ac.id/id/eprint/5165

Actions (login required)

View Item View Item
Repository STPN is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.