Introduction

Short-term rentals have reshaped cities like Nashville. This analysis used data from InsideAirbnb.com to extract insights and develop predictive models to help hosts and platforms make informed decisions. The dataset included listings, reviews, calendar data, and derived features such as the top amenities. The goals were to visualize trends and build models to predict key variables.

Data Overview

Key datasets comprised:

• Listings data — detailed information about each property
• Reviews data — reflecting guest feedback
• Calendar data — availability and pricing

These datasets were processed to extract features such as review scores, amenities, and pricing trends.

Visualizations and Exploratory Analysis

Visualizations revealed that certain neighborhoods, especially downtown Nashville, had higher concentrations of Airbnb listings. Price trends showed that neighborhoods like The Gulch commanded higher average nightly rates than East Nashville. Review scores were generally high (above 4.5), although listings with fewer amenities had slightly lower ratings.

Regression Model for Price Prediction

A linear regression model was built to predict nightly prices using features such as:

• Number of bedrooms
• Amenities
• Location
• Review scores

After applying ridge and lasso regularization, the model achieved an R² score of 0.78. Important predictors included location, number of bedrooms, and premium amenities like swimming pools and rooftop access.

Classification Model for High vs. Low Ratings

A random forest classifier predicted whether a listing would receive high ratings (above 4.5). The model achieved:

• 85% accuracy
• 0.88 precision score

Significant features included cleanliness scores, communication scores, and number of reviews.

Insights and Recommendations

The analysis produced recommendations for hosts and the platform:

• Focus on premium amenities to command higher prices and improve reviews
• Prioritize cleanliness and communication to maintain high ratings
• Highlight neighborhood advantages while differentiating listings in less premium areas

Conclusion

The Nashville Airbnb data dive provided valuable insights into factors that influence prices and guest satisfaction. Hosts can apply these findings to enhance their listings, and platforms like InsideAirbnb.com can use predictive models to offer tailored recommendations.