Abstract
In this work, we analyze more than two million news photos published in January 2016. We demonstrate i) which objects appear the most in news photos; ii) what the sentiments of news photos are; iii) whether the sentiment of news photos is aligned with the tone of the text; iv) how gender is treated; and v) how differently political candidates are portrayed. To our best knowledge, this is the first large-scale study of news photo contents using deep learning-based vision APIs.
Original language | English |
---|---|
Title of host publication | WS-16-16 |
Subtitle of host publication | CitiLab; WS-16-17: Wiki; WS-16-18: News and Public Opinion; WS-16-19: Social Media in the Newroom; WS-16-20: Social Web for Envioronmental and Ecological Monitoring |
Publisher | AI Access Foundation |
Pages | 99-107 |
Number of pages | 9 |
Volume | WS-16-16 - WS-16-20 |
ISBN (Electronic) | 9781577357681 |
Publication status | Published - 1 Jan 2016 |
Event | 10th International AAAI Conference on Web and Social Media, ICWSM 2016 - Cologne, Germany Duration: 17 May 2016 → 20 May 2016 |
Other
Other | 10th International AAAI Conference on Web and Social Media, ICWSM 2016 |
---|---|
Country | Germany |
City | Cologne |
Period | 17/5/16 → 20/5/16 |
Fingerprint
ASJC Scopus subject areas
- Engineering(all)
Cite this
Revealing the hidden patterns of news photos : Analysis of millions of news photos through GDELT & deep learning-based vision APIs. / Kwak, Haewoon; An, Jisun.
WS-16-16: CitiLab; WS-16-17: Wiki; WS-16-18: News and Public Opinion; WS-16-19: Social Media in the Newroom; WS-16-20: Social Web for Envioronmental and Ecological Monitoring. Vol. WS-16-16 - WS-16-20 AI Access Foundation, 2016. p. 99-107.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Revealing the hidden patterns of news photos
T2 - Analysis of millions of news photos through GDELT & deep learning-based vision APIs
AU - Kwak, Haewoon
AU - An, Jisun
PY - 2016/1/1
Y1 - 2016/1/1
N2 - In this work, we analyze more than two million news photos published in January 2016. We demonstrate i) which objects appear the most in news photos; ii) what the sentiments of news photos are; iii) whether the sentiment of news photos is aligned with the tone of the text; iv) how gender is treated; and v) how differently political candidates are portrayed. To our best knowledge, this is the first large-scale study of news photo contents using deep learning-based vision APIs.
AB - In this work, we analyze more than two million news photos published in January 2016. We demonstrate i) which objects appear the most in news photos; ii) what the sentiments of news photos are; iii) whether the sentiment of news photos is aligned with the tone of the text; iv) how gender is treated; and v) how differently political candidates are portrayed. To our best knowledge, this is the first large-scale study of news photo contents using deep learning-based vision APIs.
UR - http://www.scopus.com/inward/record.url?scp=85021888963&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85021888963&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85021888963
VL - WS-16-16 - WS-16-20
SP - 99
EP - 107
BT - WS-16-16
PB - AI Access Foundation
ER -