SLIDE 1
Messenger apps popularity
WebSci 2019
2
App Total users WhatsApp 1.6 bi Facebook Messenger 1.3 bi WeChat 1.1 bi Line 700 mi Telegram 200 mi
IN WHATSAPP GROUPS Josemar Alves Caetano, Gabriel Magno, Marcos - - PowerPoint PPT Presentation
IN WHATSAPP GROUPS Josemar Alves Caetano, Gabriel Magno, Marcos - - PowerPoint PPT Presentation
CHARACTERIZING ATTENTION CASCADES IN WHATSAPP GROUPS Josemar Alves Caetano, Gabriel Magno, Marcos Gonalves, Jussara Almeida, Humberto T. Marques-Neto, Virglio Almeida WebSci 2019 2 Messenger apps popularity App Total users WhatsApp 1.6
SLIDE 2
SLIDE 3
WhatsApp, politics and misinformation
WebSci 2019
3
May 2018 October 2018
SLIDE 4
What are the ingredients of the WhatsApp architecture that makes it a weaponizable social media platform to distribute misinformation?
WebSci 2019
4
Initial question
SLIDE 5
Information cascades and attention economy
WebSci 2019
5
SLIDE 6
Attention cascade example
WebSci 2019
6
SLIDE 7
WebSci 2019
7
Research questions
- How different are attention cascades in political and nonpolitical groups on
WhatsApp?
- What is the impact of false information on the characteristics of attention
cascades?
SLIDE 8
Studying attention cascades on WhatsApp using 3 dimensions
WebSci 2019
8
Structural
- Number of
nodes
- Depth
- Maximum
breadth
- Structural
Virality Temporal
- Duration
- Structural
characteristics
- ver time
User participation
- Unique users
- Relationships
motifs
SLIDE 9
Data collection and anonymization
WebSci 2019
9
Find Brazilian group links on the Internet
1 2 3
Join groups and extract messages Anonymize dataset
SLIDE 10
WhatsApp dataset
# of messages 1.7 mi # of users 31 k # of groups 120
WebSci 2019
10
SLIDE 11
Group and cascade classification
WebSci 2019
11
Cascade Political With Falsehood Unclassified Non- political With Falsehood Unclassified
Group Political Non- political
SLIDE 12
We used six Brazilian fact-checking websites
WebSci 2019
12
SLIDE 13
Overall cascade analysis
WebSci 2019
13
SLIDE 14
Structural characteristics (I)
WebSci 2019
14
SLIDE 15
Structural characteristics (II)
WebSci 2019
15
SLIDE 16
Temporal characteristics
WebSci 2019
16
SLIDE 17
Cascade duration and unique users
WebSci 2019
17
SLIDE 18
User participation motifs
WebSci 2019
18
SLIDE 19
Present and future work
WebSci 2019
19
- What kind of content would extend the duration of attention cascades?
- What is an adequate metric for quantifying attention in a group chat?
- What is the impact of hate content on the characteristics of attention
cascades?
SLIDE 20
Thank you!
WebSci 2019
20