| 摘要: | 2019年開始因為嚴重特殊傳染性肺炎影響，國與國之間開始限制出入境人數，導致全球的旅遊活動紛紛暫停，儘管有些地區疫情趨緩，也變成以「旅遊泡泡」的特殊方式方能進行交流，並僅開放部分人士出入境。
台灣是一個比起其他國家民眾更喜歡將房地產交易視為投資的地方，是否旅遊相關的因子真的能夠影響房價，進而推動不動產價格的變化，從一個較少人討論過的觀點切入這些要素影響房價的準確度，因近年來疫情影響而更值得討論。故想以國內變化為主，利用近幾年國內不同地區觀光相關的景點人數，與此次疫情染疫人數，和各縣市房價因子做資料探勘。
「食、衣、住、行、育、樂」皆為民生之本，安排旅遊已經變成台灣人中生活的小確幸，因而本研究採取六大民生事項之一：樂，較貼近人們生活中會注意到的旅遊業景氣變化一同作為房價變化預測的變數，觀察與探討由旅遊業的景氣變化預測房價的準確率，當民眾生活中有遇到相關旅遊景點人數的變化，能夠將變化納入投資房市的參考之一。
本研究利用房地產本身因子加上旅遊與疫情人數變化，透過資料探勘比較與不同的演算法預測房價預測的準確結果。以不同的回歸演算法預測，包含決策樹、支持向量機、隨機森林、類神經網路，期望本研究預測結果能作為較無經驗的購屋者搭配生活周遭變化作為買房參考。;Beginning in 2019, the COVID-19 pandemic caused a global halt to travel activities due to countries began restricting entry and exit. Even some regions saw a decline in COVID cases, travel bubbles were established as a way to facilitate communication, though, only certain individuals were permitted to enter and exit.
As Taiwan is an island nation, the tourism industry has a significant impact of the country′s income. Despite the restrictions on the tourism industry, the real estate industry has been thriving in recent years, with soaring property prices. The government has even had to introduce anti-housing speculation policies, and some have suggested that people are buying homes because they have nowhere else to put their money.
Taiwan is a place where people often invest in real estate, and it is worth considering whether tourism-related factors can truly affect property prices and be capitalized on in the real estate market. With the recent pandemic′s impact, this topic is more pertinent than ever. Therefore, this study focuses on domestic changes, and examines the factors of the number of visitors to various tourist attractions, the number of individuals infected with the virus, and housing prices in different regions of Taiwan in recent years.
"Food, clothing, housing, transportation, education, and entertainment" are all the foundations of people′s livelihood and arranging travel has become one of the little things that Taiwanese people enjoy. Therefore, this study focuses on one of the six major livelihood issues: entertainment. Taking the changes in the prosperity of the tourism industry that are closer to people′s lives as the variables for predicting changes in house prices. Observing and discussing the accuracy of predicting housing prices from the prosperity changes of the Tourism Industry. When people encounter changes in the number of visitors to various tourist attractions, they can incorporate the changes into one of their references for investing in the real estate market.
This study uses the construction factors of the real estate itself and the changes in the number of tourists and epidemics to compare and predict the accurate results of different algorithms for predicting housing prices. Using various regression algorithms, including decision trees, support vector machines, random forests, and artificial neural networks. It is hoped that the prediction results of this study can be used as a reference for inexperienced house buyers to refer to the changes in their lives when buying a property. |
