SCNARC researcher uses social media to predict personality types


Users reveal a lot about themselves through their social media profiles. Within that, they also provide insights into their personality. In this study, SCNARC researcher Jen Golbeck asked users to take a standard personality test, gathered the public data available on their Facebook profile, and showed that with machine learning techniques it is possible predict their personality scores to within roughly 10%. This work has implications for how personalty insights can be used in interface design, team building, and understanding individuals by providing a mechanism to estimate their personality from publicly accessible data. The work published in Proceedings of CHI 2011 Conference on Human Factors in Computing Systems is presented this week in Vancouver, BC [PDF].