Twitter neighborhood data

We utilized Twitter’s Streaming Application Programming Interface (API) to continuously collect a random 1% subset of publicly available geo-located. In total we have collected over 80 million tweets from over 600,000 Twitter users. Geotagged tweets were spatially mapped to the 2010 census tract and zip code areas they fall within, which enabled further assessment of the associations between Twitter-derived national neighborhood variables and neighborhood demographic, economic, business, and health characteristics. Machine-labeled and manually-labeled tweets had a high level of accuracy: 78% for happiness, 83% for food and 85% for physical activity for dichotomized labels, with the following F-scores: 0.86 (food), (0.90 (exercise), and 0.54 (happiness). <br>