Presentation for the research paper titled "An Army of Me: Sockpuppets in Online Discussion Communities" from World Wide Web conference 2017 in Perth, Australia. Research conducted by Srijan Kumar, Justin Cheng, Jure Leskovec and VS Subrahmanian from University of Maryland and Stanford University.
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
sockpuppet-www2017
1. An Army of Me: Sockpuppets in Online
Discussion Communities
Srijan Kumar Univ. of Maryland
Justin Cheng Stanford Univ.
Jure Leskovec Stanford Univ.
V.S. Subrahmanian Univ. of Maryland
4. 4
Why are sockpuppets created?
Diversify identity
Anonymize identity
Multiply identity
Gilbert, et al. (CHB, 2014), Caspi et al. (C&B, 2006)
Privacy concern
5. 5
This talk
How do sockpuppets differ from non-sockpuppets?
How are sockpuppets used?
• Are they used for deception?
• Do they support one another?
How to detect sockpuppets?
10. 10
Eric_17 April 28 2013, 12AM
Thanks. I knew Marvel fans would try to flame me, but they
have nothing other than “oh that’s your opinion” instead of
coming up with their own argument
Fellstrike April 29 2013, 6PM
Quit talking to yourself, *******. Get back on your
meds if you’re going to do that
bdiaz209 April 28 2013, 11PM
Possibly the best blog I’ve ever read major props to you
bdiaz209 posts only on this discussion to
support and defend Eric_17
Sockpuppetry in practice
11. How to define sockpuppets?
● Similar login time
● Similar login IP address
● Similar usernames
11
● Write similar to each other
● Similar point of view
● Support one another
Liu et al. (FCS 2016)
Zheng et al. (IIH-MSP 2011)
Bu et al. (KBS 2013)
These definitions make assumptions about usernames, point
of view and support
12. Defining sockpuppets
12
Sockpuppets are accounts that post from the
same IP address in the same discussion very close in
time T, in at least K different instances.
We define sockpuppets as:
3,656
Sockpuppets
1,623
Puppetmasters
T = 15 minutes, K = 3
Sockpuppets are accounts that post from the
same IP address in the same discussion very close in
time (15 min), in at least 3 different instances.
14. How do we compare sockpuppets and ordinary users?
14
For each sockpuppet, match an
ordinary user that makes
similar number of posts
on
similar discussions
We have to match!
16. Falcon-X32 Feb 5 2013, 3PM
I agree. You are absolutely right!I
16
jakey008 Feb 5 2013, 2PM
should have read the reviews first :(
ricobeans27 Feb 5 2013, 3PM
Couldn’t agree more!!
Write more self-centered posts
p < 10-3
You
Address others directly
p < 10-3
Start fewer discussions
p < 10-3
Agree more
p < 10-3
Write shorter sentences
p < 10-3
Downvoted more
p < 10-3
How do sockpuppets write?
17. 17
Smoothzilla Feb 5 2013, 3PM
Thanks for your support!!!!
Falcon-X32 Feb 5 2013, 3PM
I agree. You are absolutely right!
jakey008 Feb 5 2013, 2PM
should have read the reviews first :(
ricobeans27 Feb 5 2013, 3PM
Couldn’t agree more.
Interact more with each other
p < 10-3
Upvote each other more
p < 10-3
Relation between pair of sockpuppets
18. Do puppetmasters lead double lives?
18
Double life hypothesis:
Puppetmaster maintains distinct personality for
the two sockpuppets
More simiar Less similar
Ordinary Sockpuppet 1 Sockpuppet 2
Similarity is measured as cosine similarity between user posts’
features: LIWC, sentiment, number of words, etc.
19. Do puppetmasters lead double lives?
19
Alternate hypothesis:
Puppetmaster operates both sockpuppets
similarly
Less simiar More similar
Ordinary Sockpuppet 1 Sockpuppet 2
Similarity is measured as cosine similarity between user posts’
features: LIWC, sentiment, number of words, etc.
20. Do puppetmasters lead double lives?
20
Both sockpuppets are more similar to
each other
p < 10-3
“Good sock/Bad sock” not common
Non-sockpuppet Sockpuppet 1 Sockpuppet 2
22. Deceptiveness
22
Levenshtein distance between usernames
Numberofpairs
0 5 10 15 20
0100200300
Non-Pretenders Pretenders
Sock pairs Random pairs
srijan srijan2 srijan theRealBatman
2/31/3
Hypothesis: Deceptive sockpuppets of the same master have very different usernames.
23. 23
srijan Feb 5 2013, 3PM
i agree.. these morons dont know a thing
theRealBatman Feb 5 2013, 3PM
YOU ARE STUPID AND A *****
srijan Feb 5 2013, 2PM
best article i have read!!!
ricobeans27 Feb 5 2013, 3PM
But this article doesn’t make any sense
More opinionated
p < 10-3
Swear more
p < 10-3
Downvoted and
reported more
p < 10-3
Pretender vs non-pretender sockpuppets
25. Neutral sockpuppets
25
60%
Neutral
theRealBatman Feb 5 2013, 3PM
why so?
srijan Feb 5 2013, 3PM
best article ever!
We quantify the amount of support by counting assenting, negation
and dissenting words from LIWC
30. 30
Features
Post
Number of words,
characters, etc.,
LIWC counts,
Readability,
Sentiment,
…
Community
Number of upvotes and
downvotes,
Fraction of reported posts,
Is account reported,
…
Activity
Number of posts,
number of replies,
reciprocity of posts,
age of account,
…
Note: we are not using the IP based features
34. Do two accounts belong to the same person?
34
0.5 0.6 0.8 1.00.7 0.9
0.80
0.56
0.86
0.91
AUC
Post
Community
Activity
All
Baseline
35. An Army of Me: Sockpuppets in
Online Discussion Communities
35
Characteristics
of sockpuppets
Types of
sockpuppetry
Detecting
sockpuppets
Sockpuppets write
deceptively and
interact more
Sockpuppets very
in being deceptive
and supportive
Posts and activities
help identify
sockpuppets
Paper: bit.ly/sockpaper
Contact: srijan@cs.umd.edu
Notas do Editor
ILLUSTRATION BY SEAN MCCABE (SOURCE IMAGES: DORLING KINDERSLEY / GETTY IMAGES, SIEDE PREIS / GETTY IMAGES)
Discussions are an integral part of several web platforms, where people express their ideas, opinions and information.
Editors on Wikipedia discuss about the articles they edit; on amazon, people discuss about the products they have purchased or want to purchase; on CNN about the news, and about other things on other platforms.
However, as it is practically trivial to create accounts on these platforms, some people create more than one account to sway discussions.
These are called ‘sockpuppets’.
In fact, sockpuppets are everywhere.
For example,
On Wikipedia, people have used several sockpuppets, in some cases even hundreds, to vandalise articles and add promotional content.
On Amazon, authors have used sockpuppets to give glowing reviews to their own books.
On Yahoo discussions, CEOs have used sockpuppets to promote their own company’s stocks.
http://www.bbc.com/news/technology-34127466
http://www.nytimes.com/2007/07/16/technology/16blog.html?pagewanted=all
https://make.wordpress.org/plugins/2015/12/14/sock-puppetry-is-not-welcome/
https://www.dailydot.com/irl/wikipedia-sockpuppet-investigation-largest-network-history-wiki-pr/
https://www.raisethehammer.org/blog/1529/notice:_sock_puppet_accounts_will_be_deleted
Gilbert: Survey of 24 participants from this paper: http://www.sciencedirect.com/science/article/pii/S0747563213004287
and 104 participants from previous paper: https://link.springer.com/chapter/10.1007%2F978-0-85729-361-9_11
Caspi: 257 respondents