Anyone trying to survive in the modern world—with all of its technological bells and whistles—understands the convenience of mobile phones. And, along with it, there should also be great awareness of the benefits and risks of such a convenient piece of powerful technology. Of course, having the ability to communicate at your fingertips anywhere and anytime might make it easier to reach people when you need to, but it might also have taken some of the charm out of a simple conversation.
At the same time, though, anyone on the dating scene might also be aware of the dangers mobile phones pose after a night on the town. Indeed, cell phones have made the “drunk dial” a common fear among the socially-active community.
And as time progresses—and technology with it—some have taken to texting or even tweeting while intoxicated, only coming to regret it the morning after.
Fortunately, while technology might be partially responsible for this lackadaisical behavior, it may also provide a remedy. Apparently, machine learning now has the ability to detect alcohol-related patterns of behavior that can prevent you from sending that text in a moment-of-drunken-weakness.
It is a new kind of algorithm not necessarily ready for public use, but the University of Rochester team heading up the research say, in a new paper, “Our future work will perform a comprehensive study of alcohol consumption in social media around features such as user demographics, settings people go to drink-and-tweet. We can explore the social network of drinkers to find out how social interactions and peer pressure in social media influence the tendency to reference drinking.”
The research team collaborated with Amazon’s Mechanical Turk crowdsourcing service to collect and analyze data from 11,000 alcohol-related tweets. From this analysis, they were able to determine whether a person was simply tweeting about alcohol or if their tweets were induced by the substance. Furthermore, the research team was able to use location services to determine if these people were tweeting from home or if they were tweeting with 100 meters of their home; with as much as 80 percent accuracy.