YeeZzzy does it: Using Kanye West’s tweets to identify sleep and emotional disturbances through digital rest-activity rhythms analysis

One of the greatest challenges faced by precision medicine is the identification of biomarkers capable of detecting clinically meaningful change at the individual level, not just among large-scale population studies. To this end, the high-volume nature of an individual’s social media data could be leveraged with single-user precision to monitor sleep patterns and tweet content to determine emotional state. However, there is a lack of established methods to detect and estimate sleep and mood using social-media activity. We present here a new approach (digital rest-activity rhythms analysis) to using social media to track both sleep and mood, with potential applications to mental health monitoring and prevention. Our proof-of-concept showed that the emotional content of a single user’s tweets (Kanye West “@Ye”) were influenced by sleep disturbances inferred from usage over a 2-year period. We herein provide an ethical and theoretical-framework of how to proceed among this sensitive yet potentially fruitful field.