碩士 / 國立臺灣大學 / 建築與城鄉研究所 / 96 / This thesis attempted to simulate the land use changes by association rules of Data Mining with the visualization software NetLogo. At the meantime, the result was also compared with that by Binary Logistic Regression, the traditional method in predicting land uses. Both the simulations of these two methods were shown on NetLogo screen with five prediction accuracy indicators calculated from each result according to the real land use pattern.
At the end, these two methods have similar predicting ability of land use changes showed by the figures of accuracy indicators. But they own different strength in visual aspect. Simply black and white grids formed the building-use distribution prediction of Binary Logistic Regression which resembled more the shape of real condition in 1987. On the contrary, Association Rule can distinguish the grids into four building coverage degrees. Over all, there is still something to be done in predicting land use changes for better urban planning.
| Identifer | oai:union.ndltd.org:TW/096NTU05225009 |
| Date | January 2008 |
| Creators | Hsiao-Chun Kao, 高小鈞 |
| Contributors | 林峰田 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 119 |
Page generated in 0.0019 seconds