Online records may help unveil personal traits and attributes

Online records may help unveil personal traits and attributes

In an article entitled, “Private traits and attributes are predictable from digital records of human behavior,” published on March 11, 2013, issue of PNAS (doi:10.1073/pnas.1218772110), the lead author Michal Kosinski with two other associates affiliated to the University of Cambridge, UK & the Microsoft Research, UK have demonstrated that a variety of personal characteristics (political and religious views, gender, ethnicity, and sexual orientation) can be predicted fairly accurately from the record of a person’s  “Likes” on the Facebook. They developed a mathematical model to predict  the traits of an individual and preferences based on  the records of 58,000 U.S. Facebook users. The authors trained the model using demographic information from the volunteers’ Facebook profiles and other traits such as intelligence, personality, and satisfaction with life that were measured in online surveys and tests. The model accurately predicted the gender, ethnic origin, and sexual orientation, correctly identifying males and females in 93% of the cases, African Americans and Caucasians in 95% of the cases, and homosexual and heterosexual men in 88% of the cases under study. The model also correctly classified Democrats and Republicans as well as Christians and Muslims in more than 80% of the cases, but was less accurate at predicting relationship status, substance abuse, and parents’ relationship status. The authors further found that the model was nearly as accurate as a short personality test for predicting a user’s degree of openness to experience. The findings may be useful for improving the delivery of numerous products and services, but may also have negative implications for personal privacy. [Summarized by Samsad Razzaque a graduate student at Plant Biotech Lab. DU.]

Recent News


February 2017
Su Mo Tu We Th Fr Sa
29 30 31 1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 1 2 3