Researchers at the University of Vermont have analyzed about 43,950 Instagram images with an Artificial Intelligence capable of detecting symptoms of depression of users from the characteristics of their photographs.There were 166 volunteers who participated in this test that tried to identify related markers of said mental illness.To this end, variables such as color analysis, associated metadata and algorithmic detection of faces in photographs were taken into account.
an instagram wall full of grays
In the light of the data obtained, healthy individuals identified darker and more gray colors with a negative mood and generally preferred brighter and more vivid colors .On the contrary, it was found that depressed subjects preferred darker and more gray colors.There was also a correlation between self-identification with depression and a tendency to perceive the environment as gray or colorless. These findings led to Researchers include measurements related to color, saturation and brightness in their analysis, as well as the use of Instagram filters.
On the other hand, depression is strongly associated with the reduction of social activity. As Instagram is used to share personal experiences, it is reasonable to infer that published photos can capture aspects of social life of a user.In this premise a face detection algorithm was used to analyze the presence and number of human faces in each photograph.
The researchers also recorded the number of comments and "likes" that each message received as a measure of community participation, as well as the frequency of publication.
depression and social interaction
Photos uploaded to Instagram by depressed individuals were more blue, more gray and darker, and received less "likes." Depressed Instagram users in the sample taken for the study had a tendency to filter the color of the photos published, and showed an aversion to photos with artificial brightness, compared to non-depressed individuals.
These results coincide well with the medical literature that links depression and a preference for darker, more blue and monochromatic colors. Depressed users had more tendency to publish photos with faces, but they published fewer faces per photo. This finding may be an indicator that depressed users interact in smaller social environments.This finding would be in line with previous studies that point out that reduced social interaction is a depression indicator.
According to the study authors, Andrew G Reece and Christopher M Danforth "in the absence of a debate related to the defense of data privacy and analytical analysis, this work can serve as a model for effective detection of mental health in an increasingly digitalized society. More generally, these findings support the idea that important changes in individual psychology are transmitted in the use of social media and can be identified by computational methods ".
instagram and mental illnesses
Interestingly, the use of Instagram can already be a risk factor for mental illness, according to the study 'Social networks and mental health of young people of the Royal Public Health Society of the United Kingdom (RSPH)' This report points out that this social network can cause users insecurity related to their physical, insomnia, depression and bulling.
Comments
Post a Comment