How Your Facebook 'Likes' Could Predict Your Personality

Scientists use algorithms to show how Facebook 'likes' can accurately predict your personality

Scientists at Cambridge University have used algorithms to show how an accurate portrait of your personality could be made from the things you like on Facebook.

The research, published in journal PNAS, shows how algorithms could be used to predict your religious views, political beliefs, race and sexual orientation.

58,000 volunteers gave the researchers knowledge of their Facebook 'likes' and demographic information, these as well as test results were used to highlight personality traits.

The 'likes' were then fed into algorithms and matched with information from these tests.

The algorithms were shown to be 95 per cent accurate in differentiating African-American from Caucasian-American people, 88 per cent accurate for determining male sexuality and 85 per cent for distinguishing Republican from Democrat.

In 82 per cent of cases Christians and Muslims were correctly classified.

Substance abuse was predicted with a 65 per cent to 73 per cent accuracy.

Although the likes were rarely obviously linked to these attributes, the algorithms had aggregated large amounts of 'likes' such as music groups, food and TV programmes to create profiles.

There were also some bizarre pairings. According to David Stillwell, author of the research, curly fries correlated with high intelligence and people who liked the Dark Knight tended to have fewer Facebook friends.

However the researchers warn people could threaten their own privacy through their likes.

Michael Kosinski, lead researcher on the project, said: 'I appreciate automated book recommendations, or Facebook selecting the most relevant stories for my newsfeed. However, I can imagine situations in which the same data and technology is used to predict political views or sexual orientation, posing threats to freedom or even life.'

The researchers say users can change their privacy settings in order to make this less intrusive.

LATEST WORLD NEWS