HP Labs Research on "Twitter Cyborgs" and spam

  • 2,057 views
Uploaded on

June 2010 …

June 2010
hpl.hp.com
@hplabs

More in: Business
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
    Be the first to like this
No Downloads

Views

Total Views
2,057
On Slideshare
0
From Embeds
0
Number of Embeds
0

Actions

Shares
Downloads
0
Comments
0
Likes
0

Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide

Transcript

  • 1. TWITTER CYBORGS
    Miranda Mowbray, HP Labs
    miranda.mowbray @ hp.com
    and Nazareno Andrade, TU Delft
  • 2. “A cyborg is essentially a man-machine system in which the control mechanisms of the human portion are modified externally by drugs or regulatory devices so that the being can live in an environment different from the normal one"
    New York Times
    22 May 1960
  • 3. TWIT(TER)MINOLOGY
    Tweets
    Public timeline
    Followers
    Direct messages
    Followees
    Follow ratio
    Trend
  • 4. SOME RELATED WORK
    Java et al: follower and followee counts are correlated and have approx power-law distributions. Some automated use detected.
    A Java, X Song , T Finin and B Tseng, “Why We Twitter: Understanding Microblogging Usage and Communities”, August 2007
    Hubspot: the median number of tweets, followers, and followees are all zero.
    D Zarella, “State of the Twittersphere”, June 2009
    Cheng and Evans, Sysomos.com: 24% of all tweets made by accounts posting >150 tweets/day.
    A Cheng and M Evans, “Inside Twitter: An in-depth look inside the Twitter world”, June 2009, and “Inside Twitter: An In-Depth Look at the 5% of Most Active Users” August 2009
    Mowbray: dramatic rise in Summer ‘09 of accounts with > 100 tweets/day.
    M Mowbray, “The Twittering Machine”, April 2010 http://www.hpl.hp.com/techreports/2010/HPL-2010-54.pdf
  • 5. OUR DATA
    13447 twitterers
    Randomly sampled from public timeline between 22 April and 3 May 2010
    http://api.twitter.com/1/statuses/public_timeline.xml
    1257 spammers
    Reported to Twitter over 7 days
  • 6. RISE OF THE TWITTER CYBORGS
  • 7. RISE OF THE TWITTER CYBORGS, IIPercentage of sampled tweets that are from users sending >100 tweets/day
  • 8. “It’s about humans connecting with each other, and often in ways in which they couldn’t otherwise”
    EvWilliams
    Co-founder of Twitter
    BBC interview, 5 August 2009
  • 9. WHAT ARE THE CYBORGS DOING?
    Of the 33 sampled cyborgs tweeting > 700 times/day:
    marketing and e-commerce
    Both legit businesses and porn/phishing sites
    services for which frequent updates are useful
    Clocks, weather data, software update announcements, news headlines
    services for which they aren’t all that useful
    Torrent site, website rating site, online pastebin
  • 10. FOLLOWEE COUNTS
  • 11. FOLLOWBACK DATA
    Experiment by Catalin Cosoi
    5/10 for fake users with many tweets
    Follow ratios for sampled spammers
    0.3/10 for spammers with < 5 tweets
    4.2/10 for spammers with ≥ 5 tweets
    So if you’re a spammer, tweet at least 5 times
  • 12. THE PROBLEM WITH MEGAPHONES
    Image by Jonas Boni / daoro on Flickr, http://www.flickr.com/photos/daoro/3382798413
    CC licence http://creativecommons.org/licenses/by/2.0/deed.en_GB
  • 13. FOUR CYBORG SPAMMING TECHNIQUESand how Twitter might defend against them
  • 14. CYBORG SPAMMING 1
    Mention spam
    Mention usernames in tweets
    Tweets are automatically sent to users whose names are mentioned
    May tempt the user to follow you
    Possible defence by Twitter:
    Limit number of different usernames one user can mention in a time period
    Excluding names of their followers
  • 15. CYBORG SPAMMING 2
    Trend spam
    Automatically include trend words in your tweets
    Notoriously, #iranelection
    Tweet will be shown if a user clicks on the trend in the trends list
    Or does a Twitter-search for the trend
    Possible defences by Twitter:
    Order search results based on reputation
    Suppress multiple-trend tweets
  • 16. CYBORG SPAMMING 3
    Fake-retweet spam
    Abuse retweet convention
    “RT @name” convention for quoting others’ tweets
    Abuse this to make your ad or phishing URL appear to be from someone else
    Possible defence by Twitter:
    Track retweets, suppress “implicit” ones
    Machinery to do this is already in place
  • 17. CYBORG SPAMMING 4
    Follow spam
    Follow lots of users, send tweets or DMs to those that follow back
    Some target types of user to follow
    Some churn followees to get round Twitter’s ~2000 limit
    “Auto-Follow, Auto-Unfollow, Auto-Tweet & DM.”
    Possible defence by Twitter:
    Require a CAPTCHA solution to follow a user
    Unless the user follows you first
  • 18. A SAD STORY FROM XKCD
    Image from xkcd.com, http://xkcd.com/632 ,
    CC licence http://creativecommons.org/licenses/by-nc/2.5/
  • 19. VARIANTS
    Twitter zombies
    Used to be human, account compromised
    Eg. Barack Obama’s account
    “To increase your follower count, type your Twitter password here”
    Twitter werewolves
    Automated only at certain times, human otherwise
    Eg. Facebook games advertising on players’ Twitter accounts
    Eg. Human account owner agrees to include automated ads
  • 20. LESSONS FOR HUMAN+CYBORG ENVIRONMENTS
    Be careful when you release your API
    But do release it
    Limit automated use of the system
    Increase cost (in time, money or effort) of unwanted cyborg behaviours
    Opt-in-only marketing can still have large negative externalities
    Don’t let cyborg marketers do lots of attention-catching at small or zero cost
  • 21. Thanks