Inspiration
Tourists tend to travel along defined routes when exploring a new city, eating from the restaurants close to that route. Since location is such a critical factor for restaurant owners, distant restaurants might get less consumer traffic, damaging their sales. We wondered whether or not this trend could be proven by using the restaurant's reviews.
What it does
We investigated whether a correlation exists between a London restaurant's distance from a transportation stop (metro station or bus stop) and the number of reviews or the average review rating it received.
How we built it
We used the TourPedia API for restaurant data (latitude, longitude, review count, review polarity) and datasets for the London metro and bus system to calculate the restaurant's distance from transportation. We used Python to parse our data and create visualizations and analyses to answer our questions.
Challenges we ran into
Some challenges we ran into involved the TourPedia API. We had to figure out the exact meaning of the endpoints, parameters, and outputs to apply them to our project. Furthermore, we had some issues producing decent-quality graphs for our map-related visualizations, as there were so many data points that threw errors for the contextily Python package, which we used for the map overlay.
Accomplishments that we're proud of
We are proud of the visualizations we created, as it was the first time we worked on map overlays, heatmaps, and many of the libraries we utilized in this project. We are also proud of our results, as the distance from transportation seems to affect the number of reviews that a restaurant receives.
What we learned
We learned how to create map overlays and heat maps of different varieties, working with longitude and latitude data, and using KDTrees to find the closest transportation stop to a restaurant.
What's next for Public transportation on London restaurant reviews
The next step for our project analysis would be to eliminate confounding variables, ensuring that the data we used is indicative of a causal relationship between distance and rating.
Built With
- altair
- contextily
- python
- seaborn
Log in or sign up for Devpost to join the conversation.