Subscribe to DSC Newsletter

Contributed by Amy(Yujing) Ma. She took NYC Data Science Academy 12 week full-time Data Science Bootcamp pr...  between Jan 11th to Apr 1st, 2016. The post was based on her second class project(due at 4th week of the program). 

 

Why Airbnb?

Visiting NYC?  Airbnb is  a good choice to book unique accommodations.I have used Airbnb.com for almost 3 years, this website helps me spend my vacation as a local person, gain some fantastic experience!

To better explore its rental listings across New York City, I designed this app to answer some questions: How many of the listings are for an entire home versus a room in an apartment? How many are controlled by the same host? Why is tax Issue serious in Airbnb NYC? Should you think twice before trusting a review?

Click here to play: http://216.230.228.88:3838/bootcamp004_project/Project2-Shiny/amyma/

Data Source

Data source: Inside Airbnb

Description:  Original dataset was compiled by on 09/01/2015, containing two datasets: 1. Listings (35,957 locations and 24,426 hosts); 2. reviews (366,453 reviews).

What can we learn from the app?

How many of the listings are for an entire home versus a room in an apartment?

There are 35,957 locations and 53.6% of them are entire homes/apartments. Most of the listings are in Manhattan.

How many are controlled by the same host?

Almost 20% listings are controlled by 7% of the host.

Why is tax Issue serious in Airbnb NYC?

At the second part of Listings, Neighborhoods, and Hosts, the app would show you the top n super hosts in NYC. The table shows that most of the super hosts are not local, some of them are not even a person.

For instance, Flatbook is a company to combine hotel and Airbnb together. That's why tax issue is really serious in Airbnb NYC. People are not just sharing their places, they actually rent entire homes/apartments this as a business.

Should you think twice before trusting a review? 

Yes, be careful with those reviews contains great, nice and recommend! Based on the word cloud of reviews, people use great as frequent as stay! There are three possible reasons:

Why people tend to leave positive reviews? 

There are three possible reasons: 1. It's awkward to leave a negative one. 2. They are afraid of the hosts would leave a negative too. 3. Most of the Airbnb experience is really enjoyable.

Any suggestions for Airbnb and hosts?

1. To Airbnb, I would suggest them to send review reminder on Saturday, Sunday, and Monday. Do not send any emails during weekdays. Based on the plot, people tend to leave reviews after holidays (such as 01/02, marked as A) and on weekends (marked as B).

This plot shows that people more likely to leave reviews on Mondays and Sundays; less likely to leave one on Wednesday and Thursday since they are busy at work.

2. To hosts, I would suggest hosts choose long term rentals in February and November since the fewest people are booking during those months.

And based on the word cloud of reviews, highlight their location with some keywords: such as subway, clean, restaurant and neighborhood.

Beyond my question, the app can answer much more. To answer your questions on Airbnb in NYC, you can play with my app.

How to play?

Airbnb Listings in NYC 

To gain the basic information about Airbnb Listings in NYC, the first tab would map the whole listings.

Every circle on the map indicates one listing and different colors indicate different room types (red-Entire Home/Apt, blue- Private Room and green- Shared Room).

To get basic information on the average price in the particular neighborhood, for example, to show the average price in Manhattan, we can choose "Manhattan" on the left panel.

The right panel and map will change based on your input. In this example, I’ve selected “Manhattan".

Listings, Neighborhoods, and Hosts

The next tab shows some details about the Airbnb Listings:

To change the graphs, you can simply click the button "Make a change?"

Review by Time

The third and fourth tab would help you discover the review related to Airbnb listing in NYC.

To help users better see patterns, trends for example, in the Number of Reviews Over Time  plot, there are two methods:

Change the number in the text box at the bottom-left of the plot,  which would average the number of reviews in specified number of days. For instance, changing the number into 7, would show a smoother trend:

Go directly to the bottom plot, and show data by specified time period

Word Cloud-- Reviews

Thanks for reading

Thanks for reading, I hope you found this post and my app interesting. I want to improve this app to supply more information so if you have any suggestions please feel free to leave a comment or contact me directly.

Views: 2648

Comment

You need to be a member of Data Science Central to add comments!

Join Data Science Central

Comment by Florence Milan on May 4, 2016 at 9:50pm

Good job, analyzing Paris (France) will be interesting too.

© 2019   Data Science Central ®   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service