TY - JOUR
T1 - Predicting twitter user demographics using distant supervision from website traffic data
AU - Culotta, Aron
AU - Ravi, Nirmal Kumar
AU - Cutler, Jennifer
N1 - Publisher Copyright:
© 2016 AI Access Foundation. All rights reserved.
PY - 2016/2
Y1 - 2016/2
N2 - Understanding the demographics of users of online social networks has important applications for health, marketing, and public messaging. Whereas most prior approaches rely on a supervised learning approach, in which individual users are labeled with demo-graphics for training, we instead create a distantly labeled dataset by collecting audience measurement data for 1,500 websites (e.g., 50% of visitors to gizmodo.com are estimated to have a bachelor's degree). We then fit a regression model to predict these demographics from information about the followers of each website on Twitter. Using patterns derived both from textual content and the social network of each user, our final model produces an average held-out correlation of .77 across seven difierent variables (age, gender, education, ethnicity, income, parental status, and political preference). We then apply this model to classify individual Twitter users by ethnicity, gender, and political preference, finding performance that is surprisingly competitive with a fully supervised approach.
AB - Understanding the demographics of users of online social networks has important applications for health, marketing, and public messaging. Whereas most prior approaches rely on a supervised learning approach, in which individual users are labeled with demo-graphics for training, we instead create a distantly labeled dataset by collecting audience measurement data for 1,500 websites (e.g., 50% of visitors to gizmodo.com are estimated to have a bachelor's degree). We then fit a regression model to predict these demographics from information about the followers of each website on Twitter. Using patterns derived both from textual content and the social network of each user, our final model produces an average held-out correlation of .77 across seven difierent variables (age, gender, education, ethnicity, income, parental status, and political preference). We then apply this model to classify individual Twitter users by ethnicity, gender, and political preference, finding performance that is surprisingly competitive with a fully supervised approach.
UR - http://www.scopus.com/inward/record.url?scp=84960119619&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84960119619&partnerID=8YFLogxK
U2 - 10.1613/jair.4935
DO - 10.1613/jair.4935
M3 - Article
AN - SCOPUS:84960119619
SN - 1076-9757
VL - 55
SP - 389
EP - 408
JO - Journal of Artificial Intelligence Research
JF - Journal of Artificial Intelligence Research
ER -