本論文探討房屋內部風水對於房屋開價的三個問題，第一，房屋內部風水對其訂價是否造成影響；第二，哪些風水因素影響開價最多；第三，當被視為較不好的房屋時，是否能提出格局上的修改建議。為了研究這些問題，我們選擇使用房屋的格局圖，因為從格局圖中可以看出房屋的整體配置。由於目前沒有與房屋格局相關的資料集，根據研究目的，我們從房仲網站收集了許多房屋的格局圖及其價格以利研究。在於探討房屋內部風水與房屋訂價的關係中，我們採用了多核學習及支持向量回歸的技術來探討這些問題。其中，核心技術為使用風水書中的知識, 從格局圖中計算新的影像特徵。實驗結果顯示房屋的內部風水對於房屋訂價有影響，其中，門的位置、房屋外型與房間流通性的特徵為高度相關的風水特徵。最後，我們提出方法用於診斷風水不好的房屋。

The thesis aims at studying three questions regarding the relationship between fengshui and house pricing. First, does the internal fengshui of a house has an effect on its pricing? Second, what are the most influential fengshui features to the price? Third, can we diagnose the fengshui problems of a house? In order to answer these questions, we use house plan because a house plan shows the whole interior structure of a house. We collect numerous house plans from real estate website and their prices to facilitate our research. We use multiple kernel learning and support vector regression to study the relationship between the internal fengshui of a house and its price. All of the features are designed based on fengshui and exracted from a house plan. We found that the internal fengshui of a house has an impact on pricing. In particular, door alignment, house shape, and the room connectivity have the most significant influence. Finally, we propose a method to diagnose a house, identifying the fengshui features that may lower the house price.

[1] 謝沅瑾(2004)。《謝沅瑾居家風水教科書》。台北市：柏室科技藝術。
[2] 永慶房仲網。
[3] 〈中華民國內政部建築法〉 ( 民國100年01月05日修正 )。
[4] Der-Yang Hong, Chu-Chia Lin, “A Study on the Impact of Subway System and Road Width on the Housing Prices of Taipei”, Journal of Housing Studies, volume 8, pp. 47-67, August 1999.
[5] Chu-Chia Lin, Sue-Jing Lin, “An Analysis of the Effect of Environment Quality and Public Facilities on Housing Prices and Rents in Taiwan”, Journal of Housing Studies, volume 1, pp. 21-45, January 1993.
[6] Chu-Chia Lin, Li-Jung Huang, “An Estimation of the Impact of Unwanted Fengshui on the Prices of Commercial Real Estate in Taiwan”, Journal of Housing Studies, volume 23 no. 1, pp. 51-72, June, 2014.
[7] Rakotomamonjy, A., Bach, F., Grandvalet, Y., & Canu, S., “Simplemkl”, Journal of Machine Learning Research, Volume 9, pp. 2491–2521, 2008.
[8] C. Cortes, M. Mohri, and A. Rostamizadeh, “Learning Nonlinear Combinations of Kernels”, in NIPS 22, pp. 396–404, 2009.
[9] S. Sonnenburg, G. R¨atsch, C. Sch¨afer, and B. Sch¨olkopf, “Large Scale Multiple Kernel Learning”, Journal of Machine Learning Research, vol. 7, pp. 1531–1565, 2006.
[10] S. Vishwanathan, Z. Sun, N. Ampornpunt, and M. Varma, “Multiple Kernel Learning and the SMO Algorithm”, in NIPS, pp. 2361–2369, 2010.
[11] Ashesh Jain, S. V. N. Vishwanathan and M. Varma. “SPG-GMKL: Generalized Multiple Kernel Learning with a Million Kernels”, in Proceedings of the ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Beijing, China, August 2012.
[12] Serhat S. Bucak, Rong Jin, and Anil K. Jain, “Multiple Kernel Learning for Visual Object Recognition: A Review”, IEEE Transactions on Pattern Analysis and Machine Intelligence, volume 36, no. 7, pp. 1354-1369, July 2014.
[13] 緯來綜合台(2014)。《風水!有關係:好風水不求人》。台北市：時報文化。
[14] Dengsheng Zhang, Guojun Lu, “Shape-Based Image Retrieval Using Generic Fourier Descriptors”, International Journal of Signal Processing: Image Communication, pp. 825-848, August 2001.
[15] C. C. Chang and C. J. Lin, “Libsvm: a Library for Support Vector Machines”, ACM Transactions on Intelligent Systems and Technology, volume 2, issue 3, 2011.