Have you ever wondered just how familiar Facebook is with you?
With just 10 Facebook likes a computer algorithm can accurately profile your personality more accurately than a colleague. And with just 65 likes it can beat a roommate or friend. You might be surprised that Facebook might even know you BETTER than your family knows you.
Yes, it’s true. You probably post, like and comment on things that interest you but these are not necessarily things that you share with your family.
So how long does it take a computer to reach this level of intimacy with you? Not that long and on average we have enough ‘likes’ to support the learning!
Consider what you ‘like’ on Facebook. A study from Cambridge University and Stanford University published in the journal PNAS (Proceedings of the National Academy of Sciences), explored the ability of a computer algorithm to make accurate judgments about personality, as compared with an actual human’s ability to make personality judgments. With people, the judgments were based on familiarity with the individual while the computer relied strictly on Facebook likes. Cool huh?
The computer algorithm was able to predict an individual’s personality with greater accuracy than work colleagues, friends, roommates, parents and even siblings. The ONLY exception to the computer’s superior ability was a person’s spouse, as spouses (especially long-term) tend to know each other “inside and out.”
The average Facebook user has about 227 likes on their profile so it’s no surprise that there’s quite a lot of data for a computer to analyse, and it appears that the computer “knows” more about the individual than friends and family!
Personality traits such as openness, conscientiousness, extraversion, agreeableness, and neuroticism got analysed amongst 86,000 participants. Certain likes corresponded with particular traits, for example, ‘liking’ meditation or Salvador Dali or TED talks signified openness.
Then, friends and family (listed as friends/family on Facebook) were invited to assess the individual using a 10-point questionnaire.
How many Likes did the computer require to make the analysis?
Just 20 likes were needed for a primary analysis, with a complete personality picture emerging between 90 and 100 likes. Peak analysis performance by the computer emerged with individuals whose Likes were more prolific (250-500 likes).
So how is it that a computer is a better predictor or personality than friends or family? Dr Michal Kosinski, the study’s co-author and researcher at Stanford University, says “Machines possess several key advantages over humans: the ability to retain and access vast quantities of information nearly instantly as well as the ability to analyse it with algorithms.” Computers are not swayed by emotion, unlike humans who don’t necessarily think rationally (especially about someone they are close to).
What does this all mean?
There is an obvious opportunity for commercial business advantage in all this, especially for Marketers. Our personality influences what we ‘like’ so by understanding the 5 personality traits and what influences them, we can influence buyer behaviour.
It doesn’t stop there… Personality assessments such as this – automated, inexpensive and accurate – can be used in recruitment and dating, where the individuals best suited to the position (or potential romantic partner) are picked out far more quickly and efficiently by computers than by humans alone. Recruiters and romance aside, the marketplace – across all industries – could adjust offerings based on their users’ personalities and moods (which can be changeable).
Co-author and Cambridge researcher Dr David Stillwell said, “The ability to judge personality is an essential component of social living–from day-to-day decisions to long-term plans such as whom to marry, trust, hire, or elect as president. The results of such data analysis can be very useful in aiding people when making decisions.”
The researchers believe that such data-driven decisions can improve and simplify decision-making.
When it comes to privacy, such data mining has many people concerned. The researchers agree. Dr Kosinski stated, “We hope that consumers, technology developers, and policy-makers will tackle those challenges by supporting privacy-protecting laws and technologies, and giving the users full control over their digital footprints,” said Kosinski.