TY - JOUR
T1 - Predicting individual well-being through the language of social media
AU - Schwartz, H. Andrew
AU - Sap, Maarten
AU - Kern, Margaret L.
AU - Eichstaedt, Johannes C.
AU - Kapelner, Adam
AU - Agrawal, Megha
AU - Blanco, Eduardo
AU - Dziurzynski, Lukasz
AU - Park, Gregory
AU - Stillwell, David
AU - Kosinski, Michal
AU - Seligman, Martin E.P.
AU - Ungar, Lyle H.
N1 - Publisher Copyright:
© 2016, World Scientific Publishing Co. Pte Ltd. All rights reserved.
PY - 2016
Y1 - 2016
N2 - We present the task of predicting individual well-being, as measured by a life satisfaction scale, through the language people use on social media. Well-being, which encompasses much more than emotion and mood, is linked with good mental and physical health. The ability to quickly and accurately assess it can supplement multi-million dollar national surveys as well as promote whole body health. Through crowd-sourced ratings of tweets and Facebook status updates, we create message-level predictive models for multiple components of well-being. However, well-being is ultimately attributed to people, so we perform an additional evaluation at the user-level, finding that a multi-level cascaded model, using both message-level predictions and user-level features, performs best and outperforms popular lexicon-based happiness models. Finally, we suggest that analyses of language go beyond prediction by identifying the language that characterizes well-being.
AB - We present the task of predicting individual well-being, as measured by a life satisfaction scale, through the language people use on social media. Well-being, which encompasses much more than emotion and mood, is linked with good mental and physical health. The ability to quickly and accurately assess it can supplement multi-million dollar national surveys as well as promote whole body health. Through crowd-sourced ratings of tweets and Facebook status updates, we create message-level predictive models for multiple components of well-being. However, well-being is ultimately attributed to people, so we perform an additional evaluation at the user-level, finding that a multi-level cascaded model, using both message-level predictions and user-level features, performs best and outperforms popular lexicon-based happiness models. Finally, we suggest that analyses of language go beyond prediction by identifying the language that characterizes well-being.
UR - http://www.scopus.com/inward/record.url?scp=85012165175&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85012165175&partnerID=8YFLogxK
U2 - 10.1142/9789814749411_0047
DO - 10.1142/9789814749411_0047
M3 - Conference article
C2 - 26776214
AN - SCOPUS:85012165175
SN - 2335-6928
SP - 516
EP - 527
JO - Pacific Symposium on Biocomputing
JF - Pacific Symposium on Biocomputing
T2 - 21st Pacific Symposium on Biocomputing, PSB 2016
Y2 - 4 January 2016 through 8 January 2016
ER -