View article

Instagram is an image-based social media platform for mobile devices that promotes a new form of communication and self-expression based on images and photos. Although this image-based communication has become a wide form of online self-presentation, it is still less investigated if compared to social media text content analysis of written posts. Crucially, most Instagram images that circulate on a daily base show thin, muscular, and unrealistic body ideal, creating a breeding ground for disordered eating. Far less investigated is the emotional profile of pictures with hashtags related to eating disorders. The aim of the study was to examine the emotional expression of hashtags related to eating disorders of images posted by users on Instagram. Two hundred and fifty Instagram photos of females were considered. The hashtags were selected accordingly to a web-ranking on the most popular hashtags for eating disorders on Instagram:# anorexia,# thinispiration,# eatingdisorder,# fitinspiration and# body. The emotions expressed in each photo were measured using the Emotion API from Microsoft Azure Cognitive Service. The results showed that happiness intensity was significantly higher in images with# eatingdisorder and# fitispiration compared to# body, and higher levels of neutral emotion were found in images tagged with# body and# thinispiration. This exploratory study was one of the few studies focusing on discrete emotional expression of eating disorder photos using artificial intelligence technology.