Sunday, June 4, 2017

How Data Mining Facebook Messages Can Reveal Substance Abusers -

Substance abuse changes people’s patterns of behavior, and this is detectable in their social media messages, say researchers.


“A substance use disorder (SUD) is a condition in which recurrent use of substances such as alcohol, drugs and tobacco causes clinically and functionally significant impairment in an individual’s daily life.” So begin Warren Bickel from the Addition Recovery Research Center in Roanoke, Virginia, and a couple of pals, who study this condition.
Substance abuse is a serious concern. Around one in 10 Americans are sufferers. Which is why it costs the American economy more than $700 billion a year in lost productivity, crime, and health-care costs. So a better way to identify people suffering from the disorder, and those at risk of succumbing to it, would be hugely useful.

Bickel and co say they have developed just such a technique, which allows them to spot sufferers simply by looking at their social media messages such as Facebook posts. The technique even provides new insights into the way abuse of different substances influences people’s social media messages.
The new technique comes from the analysis of data collected between 2007 and 2012 as part of a project that ran on Facebook called myPersonality. Users who signed up were offered various psychometric tests and given feedback on their scores. Many also agreed to allow the data to be used for research purposes.
One of these tests asked over 13,000 users with an average age of 23 about the substances they used. In particular, it asked how often they used tobacco, alcohol, or other drugs, and assessed each participant’s level of use. The users were then divided into groups according to their level of substance abuse.
The results make for interesting reading. The team says its technique was hugely successful. “Our best models achieved 86%  for predicting tobacco use, 81% for alcohol use and 84% for drug use, all of which significantly outperformed existing methods,” say Bickel and co.
The technique also identified a wide range of keywords that people with substance abuse disorder are more likely to use in social media posts. “Swear words such as ‘fuck’ and ‘shit,’ sexual words such as ‘horny’ and ‘sex,’ words related to biological process such as ‘blood’ and ‘pain’ are positively correlated with all three types of substance use disorder,” say Bickel and co, referring to tobacco, alcohol and drug use. “In addition, female references such as ‘girl’ and ‘woman,’ prepositions, space reference words such as ‘up’ and ‘down’ are positively correlated with alcohol use, while words related to anger such as ‘hate’ and ‘kill,’ words related to health such as ‘clinic’ and ‘pill’ are positively correlated with drug use.”
That’s interesting work that immediately suggests a way to identify people who are at risk of substance use disorder—simply look at their Facebook posts and likes. “We believe social media is a promising platform for both studying SUD-related human behaviors as well as engaging the public for substance abuse prevention and screening,” say Bickel and co.


How Data Mining Facebook Messages Can Reveal Substance Abusers - MIT Technology Review

No comments: