Predicting Price of Real Estate in Taipei Using Support Vector Regression

• Main Authors: LEE, CHIA-HSIEN, 李家嫺
• Other Authors: TSAI, HSINE-JEN
• Format: Others
• Language: zh-TW
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/bd8yf9

碩士 === 輔仁大學 === 資訊管理學系碩士班 === 105 === In the past, the actual selling price of real estate in Taiwan was not open to the public. The buyer can only rely on estate agent subjective recommendations, resulting in buyers in the purchase is not easy to grasp the market, resulting in valuation gap. In order to let real estate prices transparent to the public, the legislation requires that every real estate transaction should be registered in real terms. The Taiwan government implemented the Actual Price Registration on August 1, 2012 which is an online database that lists the actual selling prices of property transactions in the country, yet previous researches have not been able to make effective use of the data and made comprehensive analyses for real-time login information. In this study, we use the data from Actual Price Registration database, consider and identify the effective influencing factors of real estate selling prices. We build a real estate valuation model based on the support vector regression. The purpose of the study is to establish a real estate valuation model to provide the fairness of the transaction and free of subjective human estimation error.