Skip to main content
Fig. 6 | BioMedical Engineering OnLine

Fig. 6

From: Combining crowd-sourcing, census data, and public review forums for real-time, high-resolution food desert estimation

Fig. 6

We use two sets of features to build the model: a census level features that are Kriged and are now geographically continuous, although temporally old and b features engineered using live data collected for each location from Yelp and Google; the second set of features are both geographically continuous and temporally up-to-date. We use these features and the best food desert labels to build our models for the area of study. (C) Emory University, reproduced under the CC BY-SA license

Back to article page