Computers are coming closer to understanding humans better than our own parents, siblings, or partners know us. It sounds like something out of a sci-fi movie, but researchers from the University of Cambridge have developed a system for computers to assess a person based off of their Facebook “likes,” and results are more accurate than those given by friends or family. Their findings were published in the journal Proceedings of the National Academy of Sciences.

"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," the study’s lead researcher Youyou Wu from Cambridge's Psychometrics Center, said in a press release. "In this context, the human-computer interactions depicted in science fiction films such as Her seem to be within our reach."

Researchers based their assessment on 86,220 Facebook users who voluntarily completed a 100-item personality questionnaire called “myPersonality” app and provided access to their “likes.” The average Facebook user has approximately 227 “likes,” and according to researchers this number is growing, which means artificial intelligence has more information to assess humans. The OCEAN model was used to categorize each user’s personality into the “big five” traits: openness, conscientiousness, extraversion, agreeableness, and neuroticism.

Why Computers Know You Better

When a Facebook user liked the artist Salvador Dali or meditation, for example, the computer processed the person and labeled them with a high degree of openness. Users were given the option to invite friends and family to assess their personality through the same “myPersonality” app. Out of the sample, 17,622 participants were judged by one other friend or family member, while an additional 14,410 participants were judged by two or more friends or family members.

Researchers were able to develop algorithms for the computer, which were based on decades of psychological studies and personality exam strategies. The computers took into consideration an individual’s “likes” on Facebook, and created an entire personality profile. It may be frightening to some and exciting to others, but the computer's assessment came closer to a person’s self-reported personality test scores than their brother's, mother's, best friend's, or even spouse's.

Facebook Personality Assessment Computers have learned how to assess human personality better than friends or family. Photo courtesy of University of Cambridge

"Big Data and machine-learning provide accuracy that the human mind has a hard time achieving, as humans tend to give too much weight to one or two examples, or lapse into non-rational ways of thinking," the study’s coauthor Michal Kosinski, a researcher at Stanford, said in a press release. "Nevertheless, the authors concede that detection of some traits might be best left to human abilities, those without digital footprints or dependent on subtle cognition."

The ability to accurately assess one’s personality lies in Big Data’s ability to analyze vast quantities of information with algorithms. These algorithms make it possible to make connections by collecting and storing time stamped information, the kind of records humans would rarely keep of their own friends or family members. Knowledge is power, and researchers are concerned Facebook and other social media platforms may have too much power to play with.

"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," Kosinski said. On the other hand, authors believe there could be a future for technology to efficiently judge personalities and abilities for certain jobs more aptly than self-assessed personality exams, such as the Myers-Briggs. Candidates could be matched more accurately, which could lead to improvements in job turnover rates and retention.  

"Recruiters could better match candidates with jobs based on their personality,” Wu said. “Products and services could adjust their behavior to best match their users' characters and changing moods. People may choose to augment their own intuitions and judgments with this kind of data analysis when making important life decisions such as choosing activities, career paths, or even romantic partners. Such data-driven decisions may well improve people's lives.”

 

Source: Kovinski M, Wu Y, Stillwell D. Proceedings of the National Academy of Sciences. 2015.