Researchers at the University of Cambridge and Stanford University say they have developed a statistical modeling program that analyzes a user's likes on Facebook to characterize their personality with an accuracy rivaling that of a spouse or close family member.
The researchers say the technique could help software interact with people in more meaningful ways than existing big data-based systems, which they say often make predictions that are narrow in their scope.
The researchers sampled Facebook pages from 86,220 volunteers, many of whom also filled out a personality survey focused on five major psychological traits: openness, conscientiousness, extraversion, agreeableness, and neuroticism. The researchers conducted several rounds of machine learning to associate the traits with additional Facebook likes.
To measure the effectiveness of the algorithms, the researchers gave questionnaires to friends and relatives of some participants. The survey results and computerized assessments then were compared with the self-assessments from the subjects.
The researchers found that with just 10 likes the program would know someone as well as a work colleague, with more than 70 likes it reached the level of a friend or roommate, and with more than 300 likes it reached the level of a spouse or close relative.
From IDG News Service
View Full Article
Abstracts Copyright © 2015 Information Inc., Bethesda, Maryland, USA
No entries found