介紹 (Introduction)

Airbnb provides an online platform for hosts to accommodate guests with short-term lodging. Guests can search for lodging using filters such as lodging type, dates, location, and price, and can search for specific types of homes, such as bed and breakfasts, unique homes, and vacation homes.

Airbnb為房東提供了一個在線平臺，可以為短期住宿的客人提供住宿。 訪客可以使用諸如住宿類型，日期，位置和價格之類的過濾器搜索住宿，還可以搜索特定類型的房屋，例如住宿加早餐旅館，獨特房屋和度假屋。

By reviewing the 2016 Seattle Airbnb Open Data, I will explore some interesting questions related to the lodging availability, pricing, and reviews. in addition I will try to predict the price of home listings based on the descriptive and non descriptive features.

通過回顧2016 Seattle Airbnb開放數據 ，我將探索一些與住宿可用性，價格和評論有關的有趣問題。 此外，我將嘗試根據描述性和非描述性功能預測房屋清單的價格。

While analyzing the data I found that 63% of the listings are one-bedroom property, 42% accommodates 2 guests, 37% has a strict cancelation policy and 30% has a flexible cancelation policy. Capitol Hill and Ballard are the most popular neighborhoods in the listings.

在分析數據時，我發現63％的房源為一居室物業，42％的客房可容納2位客人，37％的房屋實行嚴格的取消政策，30％的房屋實行靈活的取消政策。 國會山和巴拉德(Ballard)是清單中最受歡迎的街區。

一年中最繁忙的時間是西雅圖？ 價格上漲多少？ (What are the busiest times of the year to visit Seattle? By how much do prices spike?)

Summer season is more expensive among the year, June July and August are showing the three highest average price per home listing than the other months. The price keeps going from January (122 average) and reached the peak on July (152 average), costing on average over 23.7% than January.

一年中的夏季價格更高，6月，7月和8月是每個房屋掛牌價格最高的三個月。 價格從1月份開始(平ASP格為122)，并在7月份達到峰值(平ASP格為152)，比1月份平ASP格高出23.7％。

When I observed the rate of change of average price of lodging listings for each month, I discovered that the biggest rate of change occurred in June and the lowest in September. The first 7 months of the year also experienced a positive percentage rate of change and then subsequently August, September, October and November experienced a negative rate of change and the rate of change becomes positive again in December. This shows that there is a significant dip for around 4 months in the fall until December.

當我觀察到每個月房租平ASP格的變化率時，我發現最大的變化率發生在6月 ，而最低的變化發生在9月。 一年的前七個月也經歷了正百分比變化率，然后隨后的八月，九月，十月和十一月經歷了負變化率，并且變化率在12月再次變為正。 這表明秋季直到12月的4個月左右都有明顯的下降。

By analyzing the reviews data, I found that the number of home listings have been exponentially increased from 2009 to 2015 and were directly correlated with the number of visitors.

通過分析評論數據，我發現從2009年到2015年 ，房屋列表的數量呈指數增長，并且與訪客數量直接相關。

西雅圖最受歡迎的Airbnb房源是什么？ (What is the most popular Seattle neighborhood for Airbnb listings?)

By analyzing the listings data, I found that Capitol Hill and Ballard are the most popular neighborhoods in the Seattle listings, the below bar chart shows that Capitol Hill has 10.31 % Seattle listings, followed by Ballard with 6.26% of the listings.

通過分析清單數據，我發現Capitol Hill和Ballard是西雅圖清單中最受歡迎的社區，下面的條形圖顯示Capitol Hill擁有10.31％西雅圖清單，其次是Ballard，占6.26％。

我們可以預測西雅圖Airbnb房源的價格嗎？ 哪些方面與價格有很好的關聯？ (Can we predict a price of Seattle Airbnb listings? What aspects correlate well to price?)

It could be possible to predict the price of Seattle Airbnb listings, however its not as straight forward as it seems to be. For modeling of price prediction, I tried three algorithms, ‘Linear Regression’, ‘Random Forest Regressor’, and ‘Gradient Boosting Regressor’.

可以預測西雅圖Airbnb房源的價格，但是它并不像看起來那樣簡單。 為了對價格預測建模，我嘗試了三種算法：“線性回歸”，“隨機森林回歸”和“梯度提升回歸”。

Compared to other two models, Linear Regression achieved the best result this time where it gave an accuracy of 56% on the training set and 58% on our test set. This is due to the lack of historical data and the data requiring a huge amount of transformation to be more accurate.

與其他兩個模型相比，線性回歸這次獲得了最佳結果， 其訓練集的準確性為56％，測試集的準確性為58％。 這是由于缺乏歷史數據，并且數據需要大量轉換才能更準確。

Further analysis, I manage to find some factors that cloud influence the price of a listing in order of importance are:

進一步分析后，我設法找到一些因素會影響重要性，這些因素會影響上市價格：

· Number of bedrooms

·臥室數量

· Number of accommodates

·容納人數

· Number of Bathrooms

·浴室數量

· Room Type

· 房型

· Listing description

·清單說明

· Listing Neighborhood

·列出鄰居

結論 (Conclusion)

In this article, I tried to analyze the 2016 Airbnb Seattle data in order to answer the below questions:

在本文中，我試圖分析2016年Airbnb Seattle數據，以回答以下問題：

1. What are the busiest times of the year to visit Seattle? By how much do prices spike?

1.一年中最繁忙的時間是西雅圖？ 價格上漲多少？

2. Is there a general upward trend of both new Airbnb listings and total Airbnb visitors to Seattle?

2.新的Airbnb房源和西雅圖的Airbnb訪客總數是否都有總體上升趨勢？

3. What is the most populate Seattle neighborhood for Airbnb listings?

3. Airbnb房源在西雅圖人口最多的地區是什么？

4. Can we predict a price of Seattle Airbnb listings? What aspects correlate well to price?

4.我們可以預測西雅圖Airbnb房源的價格嗎？ 哪些方面與價格有很好的關聯？

To see more about this analysis, see the link to my Github available here

要了解有關此分析的更多信息，請參見此處的我的Github鏈接。