Project Purpose
Thank you for visiting this website! Initially created as a capstone for the Data Science Bootcamp and Nashville Software School, this project has grown into a tool for me to expand my understanding and practice of various methods of Data Analysis and Modeling. I hope to continue to develop this website as I expand my knowledge - check back in a few months, and you may find something totally new!
The Problem
The growth of AirBNB and other short-term rental companies has caused a great disruption in both the hospitality industry and the real estate industry. As AirBNB grows, it has deepened its push into gentrifying residential neighborhoods, raising home prices and pushing out long-term residents. Short-Term Rentals disproportionately impact gentrifying neighborhoods - according to definitions that I developed, gentrifying neighborhoods house approximately 8% of Nashvillians, while also being the location of 21% of the city's Short-Term rentals.
In my own exploration, I’ve found that AirBNB listings often misrepresent the neighborhoods they are in, preying on visitors unfamiliar with the city. I am a frequent user of AirBNB when I travel personally, but I am also aware of the disconnect that tourists can have when it comes to the consequences of their actions in another city. In order to build a sustainable future for the company, AirBNB and other short-term rental companies will need to become a part of the current housing shortage and gentrification crisis.
The Question
The negative sentiment towards AirBNB is quickly growing among city residents – but do AirBNB users share the feeling? I will be looking into user reviews to identify if there are patterns of users sharing a feeling of their AirBNB location.
The Method
The first step in solving this problem is identifying neighborhoods labeled as gentrifying. This is very much a hot-button issue, and there is no clear answer. What is most important to me, is to clearly explain how I identify these neighborhoods.
Once those neighborhoods have been identified, I will use various tools of Natural Language Processing (NLP) to clean my data, and then use Latent Dirichlet Allocation (LDA) to identify clusters of language from reviews in Nashville's neighborhoods.
Site Navigation
Take a look at how that the method I used to determine gentrifying neighborhoods
Identify common topics of user reviews through Natural Language Processing
Key takeaways for AirBNB operators and users, to help craft the ideal experience
Sources, disclaimers, and nowhere near enough "thank you's" to everyone who made this project a reality