Reasons why Amsterdam Airbnb listing is expensive

A look Into Amsterdam city Airbnb listings.

Introduction

In this post, I am using Data science, analysis and Python based on the CRISP-DM process to analyze the Airbnb data.

Why Amsterdam?

Amsterdam is the Netherlands capital and it is one of the top 10 travel destinations in Europe.

After looking into the data, I was interested in looking into attributes affecting the price so I came up with the following 4 questions:

Using the information in the calendar for Amsterdam Airbnb listings we can compare the availability of the listing to the average listing price for a given data.

Average Daily Price Compared to the Availability in Airbnb Listing

From the above figure we can observe the following:

Now if we compare the monthly availability of a listing compared to the average monthly prices we can find the most and cheapest months to visit Amsterdam.

Average Monthly Listing Price Compared to the Monthly Availability of Airbnb Listing

Conclusions from observing the previous image:

3) Most expensive and least expensive Neighbourhood in Amsterdam?

The next question that I was trying to answer is to find the neighborhood average price distribution.

Insights that can be derived from the above figure:

In order to find the attributes associated with the price of a listing. I used the K-means clustering method to segment the available listings into 10 clusterings. Each cluster will have attributes differentiating it from the rest of the clusters.

First, let’s look at the average listings price per cluster.

Cluster 5 had the highest average price for listings and cluster 1 had the lowest average price compared to the rest of the clusters. Thus I will look into 3 clusters and discuss the most distinguishing attributes in order to understand the pricing of a listing.

Cluster 5

Before discussing the results of the visualization, I want to clarify that the y-axis value can be ignored and it was only used to normalize the values for the attributes to ease the comparison.

The graph above shows the attributes differentiating cluster 5. this cluster contains mostly:

Cluster 1

Cluster 1 included:

Overall, the property type, neighborhood, and booking policy affected the listing price.

Conclusion

In this article, I looked into 2019 Amsterdam Airbnb listings data and answered questions related to the listings prices.

For the code and detailed analysis, please refer to the Github link:

Related posts:

We are happy to announce that manual moderation is now available on Now4real. Messages are ephemeral in Now4real so that they disappear automatically from the public after a predefined time (ranging…