Battery Amgoo AM523 Compatible Amgoo AM523 Li-Ion 2000mAh
Screen 2.0" Camera 2.0 MP Stick M2 Support up...
720p HD Bullet Camera 1/3" SONY 1.3MP CMOS Sensor 960P/960H with IR...
720p HD Dome Camera 1/4" OV 1.0MP CMOS Sensor 720P with IR Cut 3.6mm...
Categories : Apps
We all know that being tweeting in a party is not a recommended option, comparable to those whatsapps in late morning to a former couple with whom we have not talked much time ago. But it seems that there is more than one who does not follow the recommendations.
Researchers at the University of Rochester in New York, have created an algorithm that is able to detect the tweets that have been sent when the user have drunk, filtering all tweets of users in Monroe County that mention alcohol or related words to it, as drunk, beer, party, etc.
The study reveals that later, researchers are associated with Amazon's Mechanical Turk association to analyze about 11,000 tweets related to alcohol in order to distinguish between users who only wrote about it in the social network but were sober, or actually they were drinking while they tweeted.
Thus, the algorithm was able to determine whether these people were tweeting from home or within 100 meters with 80% of accuracy, which could show if more and more people drink alone at home while using the social network Twitter as a mode of entertainment.
The researchers suggest that the study is an important tool to understand better the research on alcohol consumption, the impact of social media and group pressure, and how alcohol consumption varies among different social groups.