This is a data science project that uses Airbnb Open Source data to analyze how Boston Airbnb hosts can earn more money from their properties? The analysis process uses the CRISP-DM method. The following business questions were answered during the data analysis:
- Are neighborhoods in high demand the least expensive neighborhoods?
- How can you become a superhost? What are the requirements to earn a superhost status?
- What factors increases the value of my listing?
The following Python libraries were used during the analysis:
• numpy
• pandas
• matplotlib
• seaborn
• statsmodels
• sklearn
- Boston Airbnb Data Project.ipynb – This the Jupyter notebook that contains the code for the analysis
- listing.csv: This is the csv file containing detailed listings data for Boston
In conclusion, the regression model suggests that the number of bedrooms and bathrooms, cleaning fee and the number of guests a host can accommodate plays an important role in determining the listing price, which is the value of the listing.
We also found out that superhosts tend to earn more money than regular hosts. The data suggested requirements to follow to qualify for a superhost, which was based on guest ratings, host availability, response rate, and acceptance rate. Lastly, most inexpensive neighborhoods in Boston were more likely to be on high demand than their expensive counterparts.
To get a detailed interpretation of my results and more information about this project. Visit my medium post here.
I used the following Kaggle resources in this project
How do you set the price for Airbnb? by Yuki Ueda
Seattle AirBnb Intro Analysis by Cliff Cheng