Private traits and attributes are predictable from digital records of human behavior
Kosinski, Stillwell and Graepel show that Facebook Likes, an easily accessible digital record of behaviour, can be used to predict a range of personal attributes automatically and accurately, including personality traits, intelligence, happiness, age and gender. The analysis uses a dataset of over fifty-eight thousand volunteers who provided their Likes, detailed demographic profiles and the results of several psychometric tests. The model reduces the dimensionality of the Likes data and enters it into logistic or linear regression to predict individual profiles.
Kosinski, Stillwell and Graepel show that Facebook Likes, an easily accessible digital record of behaviour, can be used to predict a range of personal attributes automatically and accurately, including personality traits, intelligence, happiness, age and gender. The analysis uses a dataset of over fifty-eight thousand volunteers who provided their Likes, detailed demographic profiles and the results of several psychometric tests. The model reduces the dimensionality of the Likes data and enters it into logistic or linear regression to predict individual profiles.
Written from the abstract as shown on the OpenAlex record, read 2026-11-04.
What our sources record
Recorded when this source was checked, figures included, with page references and quotation marks removed.
Kosinski, M., Stillwell, D. J., & Graepel, T. (2013). Private traits and attributes are predictable from digital records of human behavior. PNAS, 110(15), 5802-5805. doi:10.1073/pnas.1218772110.