With this project I set out to use the AirBnB property listing datasets for Seattle in 2016 to answer 3 questions and make 1 predictive machine learning model that could be beneficial for their business:

• Question 1: What is the average overall occupancy for all properties in Seattle over the course of the year?
• Question 2: Does it pay to be a Superhost? How do the occupancy, prices and reviews of Superhosts compare to normal hosts?
• Question 3: What neighborhoods have the highest occupancy rates?
• Prediction: Create a machine learning model to predict the average annual occupancy rate for a given property listing.

Below I will explain my findings for this project.

Visit the Github repo to see the data, code and notebooks used in the project.

3 Questions to Answer

The analyses of these questions are meant to be a starting place to inform deeper explorations of each topic, so we can gain further insights that we can apply to help guide the business.

Question 1: What is the overall occupancy in Seattle over the course of the year?

Are there periodic shifts in the overall AirBnB occupancy in Seattle over the course of the year and if so what does this look like? This can help the company decide when and how to run promotions of various kinds and to work with hosts to help them get the most out of these time frames.

The trends here are fairly easy to understand. **There are three distinct periods where we see a dramatic build up of reservations followed by a leveling off. **We’ll take a closer look at each region.

#data-science #machine-learning #exploratory-data-analysis #airbnb #regression