LAND VALUATION METHOD USING SPATIAL ANALYSIS AND ARTIFICIAL NEURAL NETWORK
Abstract: <br /> <br /> <br /> <br /> <br /> <br /> The aim of this study is to develop the land valuation method using spatial analysis and artificial neural network. The purpose of land valuation is to provide a credible and reliable land value at a given...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/7632 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Abstract: <br />
<br />
<br />
<br />
<br />
<br />
The aim of this study is to develop the land valuation method using spatial analysis and artificial neural network. The purpose of land valuation is to provide a credible and reliable land value at a given point in time. Multiple regression analysis (MRA) is the most widely used method for model calibration. It evaluates the linear relationship between a dependent (response) variable and several independent (predictor) variables, and estimates parameters for the independent variables based on a mathematical model. However, since the multicolinear value of the MRA parameters exceeds 10%, this method can not model land valuation problem precisely. ANN (artificial neural network) in the other hand can calibrate models that consist of both linear and nonlinear term simultaneously. Comparison of ANN and MRA approaches gives the land value modeling RMS (root mean square) error of Rp 149.320,00/m2 for the ANN method and Rp 375.650,00/m2 for the MRA method. Furthermore, the price-related differential (PRD) of land value modeling using ANN is 0,996. This PRD is close to 1,00, indicating that the land value estimation is neither regressive nor progressive. The land value modeling using spatial analysis and artificial neural network is a promising method for the land valuation activities. |
|---|