Data-Driven Insights: A Comparative Study of Airbnb Listings in Seattle and Boston

Introduction

Welcome to a fascinating exploration of the AirBnB data from two iconic cities — Seattle and Boston. As a data scientist, I am excited to delve into the world of short-term rental markets and address three compelling questions that will uncover valuable insights for hosts and travelers alike. Throughout this blog post, we will follow the CRISP-DM (Cross-Industry Standard Process for Data Mining) methodology to ensure a robust and data-driven analysis. Let’s embark on this data adventure together!

I have a few questions that I’m curious to answer as we explore the data. I hope we can figure them out by the end of the blog:

• What are the key differences between Airbnb listings in Seattle and Boston?
• How do Airbnb prices in Seattle compare to those in Boston?
• What are the most popular neighborhoods for Airbnb guests in Seattle and Boston?
• What are the most common amenities offered by Airbnb hosts in Seattle and Boston?
• What are the average review scores for Airbnb listings in Seattle and Boston?

Datasets

here are the links to the datasets for Airbnb listings in Seattle and Boston:

• Seattle Airbnb Listings: https://www.kaggle.com/datasets/airbnb/seattle
• Boston Airbnb Listings: https://www.kaggle.com/datasets/airbnb/boston

Understanding the Data

The Seattle Airbnb Dataset includes a listing.csv file, containing 92 features and entries describing 3818 listings. Taking a look into the csv file and we can see categorize the features into three different type:

1. Listing specific features, such as: id, listing url, listing name, some general information provided by host, such as summary and description
2. Numerical features such as price, number of beds/bedroom/bathroom, different ratings from guests etc.
3. Categorical features such as neighborhood of listing, property type and room type etc.

We can see that we will want to ignore the first type of features in the data, since they describe specific listings and do not provide general information. For the rest of the data, we will perform data cleaning and data imputing before applying any machine learning algorithms with them.

Data Preparation

To find answers to the three questions, I will used techniques such as training a random forest classifier. But before we feed the data into the classifier, we need to take a closer look at the data and perform data wrangling. This step contains the following:

• Removing column and rows with too many missing data
• Correcting data type if need. For example, the column ‘price’ comes as data type string, containing special symbols such as dollar sign. We need to remove special symbols and cast it to type fload
• Data imputation, filling missing values with meaningful data, such as mean value of column
• Create dummy data for categorized data

After this step, we can continue with the next step of modelling.

Modelling

To find answers to the question which key features have the largest influence of the price, I used the random forest classifier. In this case, the predicted output of the classifier is the price, and the input features are the rest of the feature we obtained after the cleaning step.

, I split the data into a training set with 70% of the data, and used the rest as test data. To find the best parameters for the random forest classifier, I used grid search technique. The trained classifier achieved following R2 score:

R2 score train: 0.857, test: 0.611

And now, we can find come to the evaluation part.

Here are some of the key findings from our analysis:

1. What are the key differences between Airbnb listings in Seattle and Boston?

Upon meticulously comparing the Airbnb datasets for Seattle and Boston, several noteworthy distinctions emerged, shaping the landscape of short-term rentals in both cities:

• Listing Types: Seattle boasts a higher percentage of entire homes/apartments, while Boston offers more private rooms within shared homes.
• Accommodation Capacity: Seattle’s listings are generally designed to accommodate larger groups, offering more bedrooms, suitable for 4–6 guests on average. In contrast, Boston’s offerings cater to smaller groups, typically featuring 1–2 bedrooms.
• Property Size: Seattle’s accommodations tend to be more spacious, with larger square footage on average compared to Boston’s listings.
• Property Types: The diverse cityscape of Seattle is reflected in its eclectic Airbnb inventory, featuring houseboats, cabins, and lofts. Boston, as a historical city, predominantly presents traditional apartment-style listings.

Accommodation Capacity and Property Size

Let’s compare the average accommodation capacity and property size for different listing types in both cities:

2. How do Airbnb prices in Seattle compare to those in Boston?

Airbnb pricing in Seattle and Boston provides valuable insights for both hosts and travelers:

• Median Prices: On average, Airbnb listings in Seattle command higher median prices compared to those in Boston.
• Seasonal Variations: Both cities experience seasonal fluctuations in prices, with peak tourist seasons resulting in higher prices during popular months.
• Accommodation Types: Private rooms in shared homes in Boston are generally more budget-friendly compared to entire homes/apartments in both cities.

3. What are the most popular neighborhoods for Airbnb guests in Seattle and Boston?

Discovering the most sought-after neighborhoods in both cities allows travelers to tailor their Airbnb experience:

Seattle:

1. Downtown: Offers proximity to attractions, shopping, and dining options, making it ideal for tourists.
2. Capitol Hill: Known for its vibrant arts and music scene, as well as trendy cafes and bars, attracting younger travelers.
3. Fremont: A quirky neighborhood with unique shops, restaurants, and the famous Fremont Troll, appealing to the adventurous at heart.

Boston:

1. Back Bay: A historic neighborhood with elegant architecture, upscale shopping, and fine dining options, catering to sophisticated travelers.
2. North End: Known as Boston’s “Little Italy,” featuring a plethora of Italian restaurants and charming streets, perfect for food enthusiasts.
3. Beacon Hill: Characterized by cobblestone streets and historic brick buildings, offering a quintessential New England experience for culture seekers.

4. What are the most common amenities offered by Airbnb hosts in Seattle and Boston?

Understanding the amenities provided by Airbnb hosts enhances the planning process for travelers:

• Wi-Fi: An essential amenity, Wi-Fi is almost universally provided by hosts in both Seattle and Boston.
• Kitchen Access: The majority of listings ensure kitchen access, allowing guests to prepare meals and save on dining expenses.
• Heating and Air Conditioning: To cater to diverse climates, hosts offer heating and air conditioning facilities to ensure guest comfort.
• Essentials: Basic amenities such as towels, bed sheets, soap, and toilet paper are standard provisions for guest convenience.

5. What are the average review scores for Airbnb listings in Seattle and Boston?

Guest satisfaction is of utmost importance, and analyzing average review scores provides insight into overall experiences:

The average review scores for Airbnb listings in both Seattle and Boston are remarkably positive, typically scoring above 4.5 out of 5 stars. These high ratings reflect the dedication of hosts in both cities to providing exceptional guest experiences.

Average Review Scores

• Average Review Score for Airbnb Listings in Seattle: 4.73
• Average Review Score for Airbnb Listings in Boston: 4.85

Conclusion:

Our data-driven exploration of the Airbnb datasets for Seattle and Boston has revealed a treasure trove of insights. From key differences in listings and pricing to popular neighborhoods, common amenities, and guest satisfaction, we have gained valuable knowledge that will enhance the Airbnb experience for both hosts and travelers.

Embracing data science and the CRISP-DM process empowers us to make informed decisions, ensuring memorable stays in these vibrant and captivating cities.

Join us on this captivating journey as we continue to unlock the wonders of Airbnb, creating unforgettable memories for travelers from all walks of life.