臺灣博碩士論文加值系統

本研究主要利用臺北市實價登錄資料，探討近年來臺北市房地產交易狀況及其特性，資料收集2012年至2017年一般購屋事件，篩選住宅型資料共69,868筆，透過地理資訊系統(Geographic Information System, GIS)將各筆交易資料的區域特性，如：物件與捷運站、醫院、學校、市場等公共設施的距離納入考慮，並結合實價登錄資料中與房地產交易相關之變數，如：每坪單價，房、廳、衛個數、車位、屋齡、建物類型（如：大樓、套房）等8個變項進行群集分析，本研究以自組織映射圖(Self Organizing Map, SOM)將資料分群，藉此探討各群特性。研究顯示平均單價最高的群集，近一半的交易為套房之建物類型，因此這一群的資料具備坪數低、房間數少、屋齡新等特性。而單價最低與次低的兩個群集，其物件分佈在臺北市的外環，距離市中心較遠。若比較兩兩群間的差異，單價第二高與單價第三高的兩群其物件的分佈位置與建物類型比例皆相似，但距離捷運較遠的群集，其建物的坪數較大、車位較多；距離捷運較近的群集，坪數則較小，由此可看出購屋選擇上的差異。

The purpose of this paper is to explore the status and characteristics of real estate transactions in Taipei in recent years by clustering method with Self Organizing Map. Based on 69,868 observations from the Actual Price Registration Data in Taipei City from 2012 to 2017, and select building features such as unit price, rooms, square footage, parking space, building's age, building type (e.g., apartment, suite), etc. In addition, through the Geographic Information System, the regional characteristics of each transaction data, such as the distance between transaction objects and MRT stations, hospitals, schools, and markets, are taken into consideration.
Studies have shown that cluster with the highest average price, nearly half of the building types are suites. Therefore, the characteristics of this cluster were small number of square footage, small number of rooms, and new buildings' age. The two clusters with the lowest unit price and the second lowest unit price are located in the outskirts of Taipei, far from the downtown area.
Comparing the similarities and differences between different cluster, the study found the distribution of transaction objects and the proportion of the building type are similar between the second highest unit price cluster and the third highest unit price cluster. Nevertheless, with the farther distance from nearest MRT station to its transaction objects, the number of square footage and parking space of the third cluster were more than second cluster. These result shows the difference in the choice of home purchase.

第一章 緒論 1
第一節 研究背景與動機 1
第二節 研究目的 1
第三節 研究範圍與限制 2
第二章 文獻探討 3
第一節 房地產特徵 3
第二節 資料探勘應用在房地產交易資料 3
第三章 研究方法與資料 7
第一節 研究方法 7
第二節 研究資料 21
第四章 資料分析與研究結果 35
第一節 整體資料特性 35
第二節 自組織映射圖分群結果 39
第三節 探討各群特性 41
第五章 結論與建議 57
第一節 結論 57
第二節 建議 58
參考文獻 61
附錄 63

中文參考文獻

王莉、王正歐（2003）。Tgsom：一種用於數據聚類的動態自組織映射神經網絡。電子與信息學報，25（3），313-319。
王道還（2004）。一八六一年四月十八日布羅卡發現大腦皮質上的「說話區」。科學發展（376），80-82。
李春長、童作君（2010）。住宅特徵價格模型之多層次分析。經濟論文叢刊，2（38），289-325。
政府資料開放維運團隊（2016）。政府資料開放品質指標初探檢自https://data.gov.tw/node/35273。
翁頌舜、連偉伶（2012）。以資料探勘及分析層級程序法探討家戶購屋之購屋決策因素。發表在第二十三屆國際資訊管理學術研討會。高雄市：國立高雄大學。
高健維（2006）。資料挖掘在房地產價格上之運用（未出版之碩士論文）。國立政治大學資訊管理學系。
張金鶚（2003）。房地產投資與市場分析:理論與實務(上篇: 房地產投資分析)。臺北：張金鶚。
張金鶚、范垂爐（1993）。房地產真實交易價格之研究。住宅學報（1），75-97。
陳珍華（2014）。巨量資料：公開資料與房仲網的房價分析（未出版之碩士論文）。國立交通大學資訊學院資訊學程。
陳樹衡、郭子文、棗厥庸（2007）。以決策樹之迴歸樹建構住宅價格模型—台灣地區之實證分析。住宅學報，16（1），1-20。

英文參考文獻

Bill, R. (1999). Grundlagen der geo-informationssysteme. Karlsruhe: Wichmann.
Chen, F., Deng, P., Wan, J., Zhang, D.,Vasilakos, A. V. & Rong, X. (2015). Data mining for the internet of things: Literature review and challenges - special issue on big data and knowledge extraction for cyber-physical systems. International Journal of Distributed Sensor Networks, 2015.
Cohen, R. A. (2011). Lateral inhibition. In J. S. Kreutzer, J. DeLuca & B. Caplan (Ed.), Encyclopedia of clinical neuropsychology, (Vol., pp. 1436-1437). New York, NY: Springer New York.
DeSieno, D. (1988). Adding a conscience to competitive learning. Paper presented at the International Conference on Neural Networks. CA, USA: San Diego.
Dolci, C., Salvini, D., Weibel, R. & Wiesmann, S. (2016). Spatial queries.
Erwin, E., Obermayer, K. & Schulten, K. J. (1992). Self-organizing maps: Ordering, convergence properties and energy functions. Journal Biological Cybernetics, 67(1), 47-55.
Kauko, T. (2003). On current neural network applications involving spatial modelling of property prices. Journal of Housing and the Built Environment, 18(2), 159-181.
Kohonen, T. (1982). Self-organized formation of topologically correct feature maps. Biological Cybernetics, 43, 59-69.
Kohonen, T. (1990). The self-organizing map. Proceedings of the IEEE, 78, 1464-1480.
Kohonen, T. (2013). Essentials of the self-organizing map. Neural Networks, 37, 52-65.
Kontrimas, V. & Verikas, A. (2011). The mass appraisal of the real estate by computational intelligence. Applied Soft Computing, 11(1), 443-448.
Mountcastle, V. B. (1957). Modality and topographic properties of single neurons of cat's somatic sensory cortex. Journal of Neurophysiology, 20, 408-434.
Peterson, S. & Flanagan, A. B. (2009). Neural network hedonic pricing models in mass real estate appraisal. Journal of Real Estate Research, 31(2), 147-164.
Rajasethupathy, K., Scime, A., Rajasethupathy, K. S. & Murray, G. R. (2009). Finding "persistent rules" : Combining association and classification results. Expert Systems with Applications, 36(3), 6019-6024.
Tay, D. P. H. & Ho, D. K. H. (1992). Artificial intelligence and the mass appraisal of residential apartments. Journal of Property Valuation and Investment, 10(2), 525-540.
Tolat, V. V. (1990). An analysis of kohonen's self-organizing maps using a system of energy functions. Biological Cybernetics, 64(2), 155-164.
Tulkki, A. (1998). Real estate investment appraisal of buildings using som. In G. Deboeck、T. Kohonen (Ed.), Visual explorations in finance with self-organizing maps, (Vol., pp. 128-140). London: Springer.
Yin, H. (2008). The self-organizing maps: Background, theories, extensions and applications. In J. Fulcher & L. C. Jain (Ed.), Computational intelligence: A compendium. Studies in computational intelligence, (Vol. 115, pp. 715-762). Berlin, Heidelberg: Springer.