BE CAREFUL WHAT YOU ‘LIKE’:
Researchers have made the scary revelation that what Facebook users choose to ‘like’ on the social network creates patterns that provide accurate snapshots of the user’s race, age, sexuality and other details. The study looked at 8 000 US Facebook users who volunteered their likes, demographic profiles and psychometric testing results. By applying an algorithm that uses Facebook ‘likes’, the study proved almost 90% accurate for differentiating males from females and 95% accurate in distinguishing African-Americans from whites. Christians and Muslims were classified in over 80% of cases, and accuracy of between 65% and 73% was achieved for relationship status and substance abuse. These results will also interest advertisers who can make good use of such information.