top of page
Writer's pictureCerin Sara Santhosh

Data Analysis and Visualization of Real Estate Property data


Data Analysis and Visualization of Real Estate Property data

With the introduction of Data mining and Data Visualization, the demand for real estate has increased rapidly as real estate professionals can now accurately determine the age, condition, and all other details of the buildings. They can use real estate data to buy real estate properties that are performing well in the market.


We have scraped data from the real estate website pararius.com, and in this blog, we will analyze the data to determine the following:

  • Rent per square meter

  • Balcony details

  • Number of room details

  • Upholstery details

  • Building type details

For this purpose, we will use the following tools


We have already scraped real estate data from a real estate website ( pararius.com) and explained the process in our previous blog. This is an extension of that blog, and in this, we will scrape property details from two different cities, Amsterdam and Rotterdam, and store them in two csv files. The files will be made available for download below.


The first step is to collect the data from properties from Amsterdam and Rotterdam for 30 days. We have to store the details, including the total price, total area, balcony details, room details, furnishing details, and building type details.


Get the data


The first step is to run the scraper for 30 days by scheduling it on the Datahut cloud-based data scraping platform and extract the data.


Clean the data using OpenRefine


We will clean the data using open refine, which is an open-source application. The data we get after scraping is messy and needs to be cleaned. We upload our csv file, drop all rows with blank values or values that cannot be processed. We apply transformations to make the data analytics-ready, and voila. We’re good to go.


Visualize the data


Using data wrapper. Datawrapper is a free tool for creating charts, maps, and tables.


We will analyze and visualize the following details

  • Price per square meter

  • Construction year of the buildings

  • Building type

  • Bedroom details

  • Furnishing details

  • Balcony details

Calculating Rent per square meter of Amsterdam


We first need the total rent of all apartments in Amsterdam. We also need to find the total area in square meters. Dividing the former by the latter, we get the rent per square meter. Rent per square meter is a good Key Performance Indicator for understanding how expensive /cheap the rent is. We found a lot of our customers taking rent per square meter/foot into consideration rather than just going with the rent data.


The rent per square meter value for each day is recorded and stored in the csv file.



After plotting the same using Data Wrapper, a free tool for data visualization, we get a line chart.


From this graph, we get to know that the rent per square meter in Amsterdam is around twenty-eight and that in Rotterdam is around nineteen. So after visualizing the data we get an idea of the rent per square meter. It means it is expensive to rent a house in Amsterdam compared to Rotterdam.


Construction Year of the building.


Knowing how old the building is is important when we are making an investment decision. Let's see the trend over all the buildings in Amsterdam and Rotterdam.


For analyzing the construction year of the buildings, we have to merge the files using open Refine and remove duplicates. Then we have to count the occurrence of the particular year, and we will be good to go.


Do the same thing for getting the construction year of buildings in Rotterdam.

We can also get the details of the construction year of each house. After performing the transformation functions and plotting the chart, we get a line chart.

Data Analysis and Visualization of Real Estate Property data

Data Analysis and Visualization of Real Estate Property data

Building type statistics in Amsterdam and Rotterdam


We have various types of buildings, such as apartments, independent houses, and other buildings in the data set. We thought it would be interesting to know the statistics of building types.


Data Analysis and Visualization of Real Estate Property data

From this chart, the count of apartments in both Amsterdam and Rotterdam is the same. The number of houses in Rotterdam is slightly more than that in Amsterdam. The count of buildings other than houses is more in Rotterdam.


Number of Bedroom details


The number of bedrooms is one of the most important factors that customers look for while purchasing a house. The reason is simple: if you are looking to buy a plot, you would prefer a three-bedroom or four-bedroom building. If you are single, then you would want a single-bedroom building.

Data Analysis and Visualization of Real Estate Property data

If you're looking for a one-bedroom apartment in the Netherlands, there are more options in Rotterdam than in Amsterdam. The two-bedroom options are almost equal, but Amsterdam has a slight edge when it comes to three-bedroom apartments. When it comes to four-bedroom apartments, though, Amsterdam is definitely the place to be.

Data Analysis and Visualization of Real Estate Property data

Data Analysis and Visualization of Real Estate Property data

Furnishing details


When you are looking for an apartment or house, one of the most important features you want to know is whether the building is furnished or not. In Amsterdam, almost every building is furnished with very few unfurnished. While in Rotterdam the count of unfurnished are more.

Data Analysis and Visualization of Real Estate Property data

Balcony Details


Another important feature when looking for a place to live is the presence of a balcony. Buyers often consider it as an important factor when looking for a home. This is mainly due to the fact that it gives them more room to breathe and relax. It is also a great way to enjoy the weather without having to go outside.

Data Analysis and Visualization of Real Estate Property data

The number of buildings with or without balconies is the same in Amsterdam. However, residents in Rotterdam prefer buildings with balconies more.


Conclusion

From the above visualizations, we can infer that,

  • Rent per square meter: Based on the rent, buyers may prefer a to rent or buy property in Rotterdam, as the price is lesser in Rotterdam than in Amsterdam.

  • Construction Year: The age of buildings are more or less similar in both Amsterdam and Rotterdam.

  • Building type: The number of apartments in both Amsterdam and Rotterdam is the same. Also, the number of houses in Rotterdam is slightly more than that in Amsterdam.

  • Bedroom count: The number of one-bedroom buildings in Rotterdam is more than in Amsterdam. This can be attributed to the fact that bachelor and working people don't require three to four-bedroom buildings, as they might be the ones who own them. The number of two and three-bedroom apartments are the same for both cities, as they are preferred mostly by families. Amsterdam has more buildings with four bedrooms than Rotterdam.

  • Rent based on the number of bedrooms: The price of buildings in Amsterdam with one room and two rooms are almost volatile. While three bedrooms and four bedrooms are highly varying. The same stands for the buildings in Rotterdam.

  • Furnishing Details: Buildings in Amsterdam are furnished with very few remaining unfurnished. However, in Rotterdam, there are a few buildings with unfurnished interiors.

  • Balcony Details: Residents in Amsterdam have the same preference for buildings with balconies and without balconies. On the other hand, residents in Rotterdam prefer buildings with balconies.


The purpose of this blog is to show you how you can make use of data on the web and make actual insights. For knowing more about how to scrape a real estate website, check out our previous blog: https://www.blog.datahut.co/post/scraping-property-data


Download the property data scraped from Amsterdam

Download the property data scraped from Rotterdam


Related Reading:


1,806 views

Do you want to offload the dull, complex, and labour-intensive web scraping task to an expert?

bottom of page