Now by the pictures in Instagram to determine the psychological state

Modern computer algorithms capable of much these days.

They know how to drive, to choose the most appropriate playlist according to the user preference of streaming services like Spotify, you can retouch photos, and even describe the world of blind people. Now that machine learning is able to look at your Instagram feed services, to analyze it and decide what psychological condition you are in and whether you have clinical depression.

Published on the SpringerOpen results of the last research refers to the analysis of 166 feeds Instagram, in which the car was looking for the so-called “markers of clinical depression”. The authors of the study claim that the assessment carried out by a computer algorithm, was even more successful than the detection and diagnosis of this condition these practitioners.

“Statistical data were collected based on the analysis of a total of 43 950 user photos from Instagram, which were evaluated for color, metadata, and the use of algorithmic recognition technique. the Results of the detection of signs of depression have exceeded the average indicators of efficiency of doctors of General medical practice.”

The algorithm was tested from a number of different factors, ranging from the color photos (including verification of use or non-use by the user different color filters) to the frequency of publications.
“In the mood, colors, and mental health revealed that more dark and grey that psychologically healthy people are often associated with low mood and more vivid with good. Our investigation has shown that people who are depressed, choose a more dark and gray tones for your photos […]”.

“Depression is closely associated with reduced social activity. Since Instagram is used for sharing your personal experience, it would be reasonable to conclude that the published pictures can display and social aspects of life of the user. Based on this, we used the algorithm of face recognition to analyze publications in Instagram to determine the presence and number of human faces on each of the photos. We also counted the number of comments and likes under each post that displays community engagement, and used the publishing frequency as an indicator of user interaction”.

© 2017 – 2019, paradox. All rights reserved.