Diagnosis on Facebook
AI diagnoses depression earlier than doctors by posts in social networks
Daria Berdnikova, Hi-tech+
If the words "tears" and "feelings" are found in your messages on social networks, complaints about loneliness, and the pronoun "I" and its forms are used many times, this is a marker of psychological problems.
Scientists from the University of Pennsylvania and the State University of New York at Stony Brook have found another way to diagnose depression. It turned out that posts in social networks give out about the patient that the doctor may not be noticed at the reception, according to a publication in the scientific journal Proceedings of the National Academy of Scientists (Eichstaedt et al., Facebook language predicts depression in medical records).
Topics and keywords in messages that showed the maximum positive correlation with the probability of developing depression (from an article in PNAS) – VM.
"Information from social networks contains markers like a genome,– explains scientist Johannes Eichstadt on the website of the University of Pennsylvania. – It's amazing that we can use similar methods to analyze social media data to find these markers. Thus, depression really changes people's activity on social networks, and in the case of skin diseases or diabetes, nothing like this is observed."
Portal "Eternal youth" http://vechnayamolodost.ru