Socioeconomic data is as valuable as it is difficult to come by. What if we can use existing data about restaurants to predict socioeconomic factors? We built a model that takes in restaurant data for 9 Chinese cities from Dianping, the Chinese equivalent to Yelp. This data ranges from cuisine type to restaurant ratings.
Using this data, we can predict factors like daytime population, nighttime population, company presence, and spending amount. These predictions come at more granular scale, both in time and space, than traditional census data. Below, we show predictions for 3km2 grid cells and compare the results and accuracy across the studied cities.
# restaurants
restaurant diversity
prediction
Daytime Population
Nighttime Population
Number of Companies
Spending
Select the cities in the map to filter the paths in the chart.
# restaurants
predicted totals
prediction accuracy
Siqi Zheng Director, MIT China Future City Lab
Carlo Ratti Director, MIT Senseable City Lab
Lei Dong Lead Researcher
Fábio Duarte Design Manager
Sarah Campbell Web, Visualization
Download press material
The material on this website can be used freely in any publication provided that:
1. It is duly credited as a project by the MIT Senseable City Lab and MIT China Future City Lab
2. A PDF copy of the publication is sent to senseable-press@mit.edu
Predicting neighborhoods’ socioeconomic attributes using restaurant data
Lei Dong, Carlo Ratti, and Siqi Zheng, 2019
Proceedings of the National Academy of Sciences of the United States of America
Support: National Science Foundation of China and MITOR
Accessibility