Measuring and Characterizing Nutritional Information of Food and Ingestion Content in Instagram

Social media sites like Instagram have emerged as popular platforms for sharing ingestion and dining experiences. However research on characterizing the nutritional information embedded in such content is limited. In this paper, we develop a computational method to extract nutritional information, specifically calorific content from Instagram food posts. Next, we explore how the community reacts specifically to healthy versus non-healthy food postings. Based on a crowdsourced approach, our method was found to detect calorific content in posts with 89% accuracy. We further show the use of Instagram as a platform where sharing of moderately healthy food content is common, and such content also receives the most support from the community.