Google Play icon

Cambridge Computer Algorithm Knows You As Good As Your Spouse

Posted January 19, 2015

These days, saying that Facebook knows a lot about us is a virtually redundant statement. Everyone is aware of the fact that it collects data about our lives, which sometimes leads to such bizarre things as losing one’s job over a slightly-less-than-wholesome picture or even getting arrested on criminal charges.

Scientists from Cambridge University fed Facebook „likes“ into a computer algorithm designed to predict people’s personality traits. Image credit: bykst via, CC0 Public Domain.

Scientists from Cambridge University fed Facebook „likes“ into a computer algorithm designed to predict people’s personality traits. Image credit: bykst via, CC0 Public Domain.

But just how well does this platform really knows us? According to a group of scientists from Cambridge University, the answer is “very well”.

They’ve come up with a computer algorithm that proved to be almost uncanny in its accurate predictions of a subject’s personality based on nothing but his or her Facebook “likes”.

In fact, the program is so accurate, that it put even friends and family members to shame.

“Computers do better than human beings in most cases,” said Youyou Wu, a PhD student in the Department of Psychology at the University of Cambridge and co-author of the study. “In some cases, the computer’s judgement can even describe real-life behaviours better than self-ratings.”

The algorithm used personality data and Facebook “likes” of over 85,000 volunteers.

The data consisted of answers to a massive, 100-item questionnaire (provided via the myPersonality app), which focuses on the Big Five personality traits in psychology: openness, conscientiousness, extraversion, agreeableness and neuroticism.

“We basically asked our computer model to look at the association between “likes” and personalities,” said Wu. “We asked the computer to make a judgement for these people, based on their self-ratings.”

To see how accurate the program is at predicting individual traits, the team asked the participants’ families and friends to fill out a shorter questionnaire. The results were fairly astounding.

As few as 10 “likes” were necessary for the algorithm to make more accurate predictions than a work colleague, and 150 “likes” outperformed a parent, sibling or partner.

Spouses (300 “likes”) were generally better than the program, ye the difference was almost negligible.

Analysis of the data showed which “likes” equated with higher levels of particular traits. For instance, liking Salvador Dali or meditation revealed a higher degree of openness, while loving the movie Silence of the Lambs and the music of Nicki Minaj were both correlates of competitiveness.

“In the future, computers could be able to infer our psychological traits and react accordingly, leading to the emergence of emotionally intelligent and socially skilled machines,” noted Wu.

“In this context, the human-computer interactions depicted in science fiction films such as Her seem to be within our reach.”

Tampering their optimism are concerns about privacy as such technology develops. For this reason, they emphasise the importance of giving users full control over their personal content, thus decreasing the likelihood of privacy invasion.

The study was published ahead of print on the 12th of January in the Proceedings of the National Academy of Sciences.

Sources: study abstract,,

Featured news from related categories:

Technology Org App
Google Play icon
85,413 science & technology articles

Most Popular Articles

  1. New treatment may reverse celiac disease (October 22, 2019)
  2. "Helical Engine" Proposed by NASA Engineer could Reach 99% the Speed of Light. But could it, really? (October 17, 2019)
  3. New Class of Painkillers Offers all the Benefits of Opioids, Minus the Side Effects and Addictiveness (October 16, 2019)
  4. The World's Energy Storage Powerhouse (November 1, 2019)
  5. Plastic waste may be headed for the microwave (October 18, 2019)

Follow us

Facebook   Twitter   Pinterest   Tumblr   RSS   Newsletter via Email