Pemodelan Nilai Tanah di Sekitar Bandara Dhoho, Kabupaten Kediri Menggunakan Algoritma Neural Network

Salsabilla, Yesha Putri (2025) Pemodelan Nilai Tanah di Sekitar Bandara Dhoho, Kabupaten Kediri Menggunakan Algoritma Neural Network. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5016211071-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (7MB) | Request a copy

Abstract

Pembangunan Bandara Dhoho di Kabupaten Kediri, Jawa Timur, berperan penting dalam meningkatkan aksesibilitas dan pertumbuhan ekonomi di wilayah sekitarnya. Salah satu dampak signifikan yang ditimbulkan adalah perubahan nilai tanah di sekitar bandara. Penelitian ini bertujuan untuk menganalisis perubahan nilai tanah sebelum dan sesudah pembangunan bandara serta memprediksi nilai tanah pada tahun 2024 menggunakan algoritma Artificial Neural Network (ANN) dan Random Forest (RF). Studi dilakukan pada area dengan radius 1 kilometer dari bandara, mencakup tujuh desa, yaitu Tarokan, Bulusari, Kalipang, Kaliboto, Grogol, Jatirejo, dan Tiron. Hasil penelitian menunjukkan bahwa seluruh zona mengalami kenaikan nilai tanah, dengan persentase peningkatan antara 18,78% hingga 89,08%. Model ANN menunjukkan performa lebih baik dibanding RF, dengan nilai Root Mean Square Error (RMSE) sebesar Rp37.124, Mean Absolute Error (MAE) sebesar Rp33.687, Mean Absolute Percentage Error (MAPE) sebesar 9,07%, dan koefisien determinasi (R²) sebesar 0,9519. Sementara itu, model RF menghasilkan RMSE sebesar Rp45.182, MAE sebesar Rp37.229, MAPE sebesar 10,23%, dan R² sebesar 0,8512. Nilai Price-Related Differential (PRD) dari ANN sebesar 0,991 menunjukkan distribusi prediksi yang merata, sedangkan PRD RF sebesar 1,109 mengindikasikan kecenderungan overestimasi pada zona bernilai tinggi. Peta prediksi nilai tanah tahun 2024 berdasarkan model terbaik, yaitu ANN menghasilkan sebaran nilai antara Rp35.358 hingga Rp915.743 per meter persegi, dengan nilai tertinggi berada di sekitar akses jalan utama di sisi selatan bandara. Penelitian ini diharapkan menjadi referensi penting dalam perencanaan tata ruang, pengelolaan lahan, serta kebijakan pajak berbasis nilai pasar tanah di kawasan sekitar Bandara Dhoho.
===================================================================================================================================
The construction of Dhoho Airport in Kediri Regency, East Java, plays a significant role in improving accessibility and stimulating economic growth in the surrounding area. One of the major impacts is the change in land value around the airport. This study aims to analyze land value changes before and after the airport development and to predict land values in 2024 using Artificial Neural Network (ANN) and Random Forest (RF) algorithms. The study area covers a 1-kilometer radius around the airport, encompassing seven villages: Tarokan, Bulusari, Kalipang, Kaliboto, Grogol, Jatirejo, and Tiron. The results indicate that all zones experienced an increase in land value, with percentage growth ranging from 18,78% to 89,08%. The ANN model outperformed the RF model, achieving a Root Mean Square Error (RMSE) of IDR 37.124, Mean Absolute Error (MAE) of IDR 33.687, Mean Absolute Percentage Error (MAPE) of 9,07%, and a coefficient of determination (R²) of 0,9519. In contrast, the RF model yielded an RMSE of IDR 45.182, MAE of IDR 37.229, MAPE of 10,23%, and R² of 0,8512. The Price-Related Differential (PRD) for ANN was 0,991, indicating evenly distributed predictions, while the RF PRD value of 1,109 suggests a tendency to overestimate high-value zones. The 2024 land value prediction map, based on the best-performing model (ANN), shows a value distribution ranging from IDR 35.358 to IDR 915.743 per square meter, with the highest values concentrated around the main access roads on the southern side of the airport. This study is expected to serve as an important reference for spatial planning, land management, and land value-based tax policy development in the area surrounding Dhoho Airport.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Nilai Tanah, Neural Network, Perubahan Nilai Tanah, Prediksi Nilai Tanah Land Value, Neural Network, Land Value Change, Land Value Prediction
Subjects:	G Geography. Anthropology. Recreation > GA Mathematical geography. Cartography > GA109.5 Multipurpose cadastres. T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models.
Divisions:	Faculty of Civil, Planning, and Geo Engineering (CIVPLAN) > Geomatics Engineering > 29202-(S1) Undergraduate Thesis
Depositing User:	Yesha Putri Salsabilla
Date Deposited:	23 Jul 2025 08:49
Last Modified:	23 Jul 2025 08:49
URI:	http://repository.its.ac.id/id/eprint/120916

Actions (login required)

View Item