Comments on: Addressing Twitter Spam Through Statistical Analysis

By: How to Deal With Low-Quality Messages

How to Deal With Low-Quality Messages — Mon, 27 Apr 2009 19:53:45 +0000

[...] about who you follow on social networks and you will reduce your intake of low-quality messages. In this great post by Damon Cortesi, he shares a strategy for identifying spammers and heavy self-promoters in order to avoid them. Are [...]

By: Stop Twitter Spam » Twitter Stumbles Again With Accidental Account Deletions

Sat, 02 Aug 2008 21:26:25 +0000

[...] of situation and they may help prod the devs with some other ideas as to how to help prevent spam. http://dcortesi.com/2008/04/16/addressing-twitter-spam-through-statistical-analysis/ [...]

By: The Tuesday Night Tech Show

The Tuesday Night Tech Show — Wed, 16 Apr 2008 16:23:22 +0000

Great technique to weed out these Twitter spammers–I hate getting 10-20 new friend requests a day only to find out most of them are garbage spammers. I have tried to start discussions on “Get Satisfaction” and I suggest you all do the same–Twitter has a company page set up for complaints with employees reading and writing back. My recent topic I started is: http://getsatisfaction.com/twitter/topics/please_stop_the_twitterbot_army_twitter_spam_must_stop

By: justin

justin — Wed, 16 Apr 2008 15:46:28 +0000

spam accounts often post a handful of tweets at almost the exact same time every day–prob’ly because the spamming scripts are run by cron jobs. they usually only use one interface–web, of course. unlike a real person, they will jump right in and start tweeting, and will continue at the same tweet volume. a real person starts out slow, and builds tweet volume as they build followers and have more people to interact with. so most spam tweeters will have very uneven tweetstats graphs, in almost every category…

By: Stop Twitter Spam » Statistical Analysis of a Twitter Spammer

Stop Twitter Spam » Statistical Analysis of a Twitter Spammer — Wed, 16 Apr 2008 14:44:13 +0000

[...] Damon Cortesi (@dacort, creator of TweetStats) has a great post today about statistical analysis of a Twitter Spammer. This is a must-read for anyone who is concerned about the rise of spam on Twitter. We’ve been looking at a lot of spammer profiles over the past few weeks, and Mr. Cortesi is absolutely correct – these profiles often have a very similar statstical “signature” in terms of their Following:Follower ratio, whether or not their updates include links back to their profile URL, etc. And he’s even looking at the percentage of Followers who followed via an auto-follow, which is a subject that we just posted about. Furnishing an easy means by which to report/block spam is also a necessary evil. Twitter has hummed along relatively under the spam radar until now, but it seems it has to accept that spammers will try to take advantage of its users. Giving users the power to identify and avoid spam through the use of statistics will hopefully make Twitter a fruitless source of successful spam. [...]

By: kapheroph

kapheroph — Wed, 16 Apr 2008 09:27:33 +0000

Thanks for the analysis, that was great. It seems like everything it does come down to statistics and probability theory. It’s going to be interesting to see where you end up with this. Keep us posted.

By: Danacea

Danacea — Wed, 16 Apr 2008 08:49:54 +0000

Nicely done. Helps all Twitterers be aware of the problem and sort the spammers from the real accounts. This is becoming a serious (yet sadly inevitable) problem and each person needs to move to change it!