Use Spatial Data Mining for Planning Urban Mass Rapid Transit System

資料探勘可藉由以往歷史資料之探索取得未來有用的資料，而都會區捷運系統之需求亦係基於大眾對公共運輸（空間）需求而衍生；因此，本研究從需求面探討相關歷史資料及大眾運輸特性確認資料探勘之範疇而取得既有公車系統、交通樞紐點及重要公共設施有其關聯性；其同質性之部份均為地理位置，故採用地理資訊系統(GIS)將該三項關聯要素之各個地點予以定位疊合，惟因該三項要素功能性質不同，續借助訪談多位交通專家確立並取得其相互間關聯強度，並將該結果以AHP分析該三項對站址選擇之權重引入GIS內，藉由群集分析得到適度之車站數量，並尋得最符合捷運運輸需求之站址；為確認本研究方法之可行性，乃以人口逾150萬人之桃園都會區為例作驗證，該都會區捷運系統已依傳統之現場調查、居民問卷及所有交通系統分析等百式規劃完成，經依本研究之資料探勘程序施作後，所獲得之捷運系統站址與傳統方式之規劃結果頗為近似；故應具有可用性。

關鍵字

資料探勘；都會區捷運系統；集群分折； GIS ；疊合分析； AHP

並列摘要

Find the related historical data and Mass Rapid Transit System (MRTS) attributes from the demand aspect to confirm the data mining range and get the relevance between factors, which have common attribute of geographic location. Therefore, we apply GTS、 AHP and clustering analysis to get locations of MRTS. Apply these methods on Taoyuan Urban Area with a population of 15 million. Its Urban MRTS has been completed by conventional analysis methods. The locations obtained by data mining are very close to the planning results by conventional methods. Thus, the data mining is useful to planning MRTS.

並列關鍵字

Data Mining ； Urban Mass Rapid Transit System ； Clustering Analysis ； GIS ； Overlay Analysis ； AHP

參考文獻

Lee, C.,Chen, C.F.,Huang, S. M.,Hsu, C. J.,Sheu, S. W.(2002).Development of an automated and integrated airport pavement image collection system.Journal of Marine Science and Technology.10(1),1-7.

Google Scholar

Chou, J. C.,Wende, A. O`Neill,Cheng, H. D.(1994).Pavement distress classification using neural networks.Systems, Man, and Cybernetics, Humans Information and Technology, IEEE International Conference.(Systems, Man, and Cybernetics, Humans Information and Technology, IEEE International Conference).