Identifying and nuking Twitter spammers

Warning: this content is older than 365 days. It may be out of date and no longer relevant.

Twitter’s the hot new thing, the shiny object du jour. As such, it’s also turned into a massive cesspool of spam from marketers desperate to try hawking their ineffective wares in another channel, hoping against hope that consumers on Twitter are not as smart at filtering them out as they are in other media.

Sorry, guys. This blog post is about making your life harder.

Here’s how to identify Twitter spammers in your personal timeline using Yahoo Pipes.

Go to Yahoo Pipes and start a new pipe. Grab a Fetch Feed box from Sources and drag it into the worksheet.

In the box, insert your Twitter personal timeline. It’s formatted like this:

https://username:[email protected]/statuses/friends_timeline.rss

where obviously username and password are your Twitter username and passwords.

Next, drag two filter boxes from Operators. Drag the blue circle at the bottom of the Fetch Feed to the first Filter box.

Then drag the blue circle from the bottom of the first Filter Box to the second, and from the bottom of the second to Pipe Output.

Set the first to Block All and the second to Permit Any.

In Block All, set the item title dropdown to @. This filters out @ replies, since those are likely to be a little more legitimate than pure crap tweets. Not much, but at least a little.

In the Permit Any filter, start adding text in for the tweets you know are garbage. Typically they have “make money” in them, words like “F*R*E*E” and other useless fare. Add these line by line until you have a list of the garbage.

Yahoo Pipes making a hit list

Name, save, and run the pipe. If all goes well, you’ll see a screen with options.

Pipes: Twitter ID Spammers

From that RSS box, you can subscribe to this Yahoo Pipe in the feed reader of your choice. All of the tweets that end up in it should be crap, which you can then promptly unfollow either manually from your feed reader or automatically if you’re handy at writing against the Twitter API.

Next, grab a beer, wait a few days for the pipe to fill up, then say farewell to people using Twitter as just another dumping ground or a meager prop for their failed business model as you unfollow them.

Did you enjoy this blog post? If so, please subscribe right now!

Identifying and nuking Twitter spammers 1 Identifying and nuking Twitter spammers 2 Identifying and nuking Twitter spammers 3

Get this and other great articles from the source at www.ChristopherSPenn.com


Comments

44 responses to “Identifying and nuking Twitter spammers”

  1. Excellent post. I always here about your use of Yahoo Pipes on Marketing Over Coffee. It is very exciting to see exactly how to make one. Keep em coming!

  2. Excellent post. I always here about your use of Yahoo Pipes on Marketing Over Coffee. It is very exciting to see exactly how to make one. Keep em coming!

  3. Excellent post. I always here about your use of Yahoo Pipes on Marketing Over Coffee. It is very exciting to see exactly how to make one. Keep em coming!

  4. Excellent post. I always here about your use of Yahoo Pipes on Marketing Over Coffee. It is very exciting to see exactly how to make one. Keep em coming!

  5. a great tip with all the detail needed to make it work – thanks for sharing this!

  6. a great tip with all the detail needed to make it work – thanks for sharing this!

  7. a great tip with all the detail needed to make it work – thanks for sharing this!

  8. a great tip with all the detail needed to make it work – thanks for sharing this!

  9. Short of running some Yahoo Pipes from code other people wrote, I don't know the animal. Like Bob wrote, thanks for the detail! How does one write against the API? ๐Ÿ™‚

  10. Short of running some Yahoo Pipes from code other people wrote, I don't know the animal. Like Bob wrote, thanks for the detail! How does one write against the API? ๐Ÿ™‚

  11. Short of running some Yahoo Pipes from code other people wrote, I don't know the animal. Like Bob wrote, thanks for the detail! How does one write against the API? ๐Ÿ™‚

  12. Short of running some Yahoo Pipes from code other people wrote, I don't know the animal. Like Bob wrote, thanks for the detail! How does one write against the API? ๐Ÿ™‚

  13. Love the concept, but I take issue with the terms you chose as the marks of a spam tweet. I'm sure I have tweeted “Whew, just finished a long new blog post!” or something of the sort … and there are perfectly legitimate people who share a blog post they wrote every once in awhile preceded by “New Blog Post:” Non-spammers could also tweet “Need some way to make more money” as a vent and not as a scam invitation.

    You can definitely use this method to find people who might throw up red flags, but I'd refrain from doing it automatically, as then it would have automatically unfollowed you, according to Summize: http://tinyurl.com/cttjsu

    Even a 100% spammy term like tweetergetter will probably come up in some tweets by people who say “I hate tweetergetter.”

    Have you seen TweetSum? They're still working on the algorithm, but it ranks a user's past behavior against a “douchebag index” and the spammers move to the top. http://www.tweetsum.com

  14. Love the concept, but I take issue with the terms you chose as the marks of a spam tweet. I'm sure I have tweeted “Whew, just finished a long new blog post!” or something of the sort … and there are perfectly legitimate people who share a blog post they wrote every once in awhile preceded by “New Blog Post:” Non-spammers could also tweet “Need some way to make more money” as a vent and not as a scam invitation.

    You can definitely use this method to find people who might throw up red flags, but I'd refrain from doing it automatically, as then it would have automatically unfollowed you, according to Summize: http://tinyurl.com/cttjsu

    Even a 100% spammy term like tweetergetter will probably come up in some tweets by people who say “I hate tweetergetter.”

    Have you seen TweetSum? They're still working on the algorithm, but it ranks a user's past behavior against a “douchebag index” and the spammers move to the top. http://www.tweetsum.com

  15. Love the concept, but I take issue with the terms you chose as the marks of a spam tweet. I'm sure I have tweeted “Whew, just finished a long new blog post!” or something of the sort … and there are perfectly legitimate people who share a blog post they wrote every once in awhile preceded by “New Blog Post:” Non-spammers could also tweet “Need some way to make more money” as a vent and not as a scam invitation.

    You can definitely use this method to find people who might throw up red flags, but I'd refrain from doing it automatically, as then it would have automatically unfollowed you, according to Summize: http://tinyurl.com/cttjsu

    Even a 100% spammy term like tweetergetter will probably come up in some tweets by people who say “I hate tweetergetter.”

    Have you seen TweetSum? They're still working on the algorithm, but it ranks a user's past behavior against a “douchebag index” and the spammers move to the top. http://www.tweetsum.com

  16. Love the concept, but I take issue with the terms you chose as the marks of a spam tweet. I'm sure I have tweeted “Whew, just finished a long new blog post!” or something of the sort … and there are perfectly legitimate people who share a blog post they wrote every once in awhile preceded by “New Blog Post:” Non-spammers could also tweet “Need some way to make more money” as a vent and not as a scam invitation.

    You can definitely use this method to find people who might throw up red flags, but I'd refrain from doing it automatically, as then it would have automatically unfollowed you, according to Summize: http://tinyurl.com/cttjsu

    Even a 100% spammy term like tweetergetter will probably come up in some tweets by people who say “I hate tweetergetter.”

    Have you seen TweetSum? They're still working on the algorithm, but it ranks a user's past behavior against a “douchebag index” and the spammers move to the top. http://www.tweetsum.com

  17. Excellent post. I always here about your use of Yahoo Pipes on Marketing Over Coffee. It is very exciting to see exactly how to make one. Keep em coming!

  18. a great tip with all the detail needed to make it work – thanks for sharing this!

  19. Short of running some Yahoo Pipes from code other people wrote, I don't know the animal. Like Bob wrote, thanks for the detail! How does one write against the API? ๐Ÿ™‚

  20. Love the concept, but I take issue with the terms you chose as the marks of a spam tweet. I'm sure I have tweeted “Whew, just finished a long new blog post!” or something of the sort … and there are perfectly legitimate people who share a blog post they wrote every once in awhile preceded by “New Blog Post:” Non-spammers could also tweet “Need some way to make more money” as a vent and not as a scam invitation.

    You can definitely use this method to find people who might throw up red flags, but I'd refrain from doing it automatically, as then it would have automatically unfollowed you, according to Summize: http://tinyurl.com/cttjsu

    Even a 100% spammy term like tweetergetter will probably come up in some tweets by people who say “I hate tweetergetter.”

    Have you seen TweetSum? They're still working on the algorithm, but it ranks a user's past behavior against a “douchebag index” and the spammers move to the top. http://www.tweetsum.com

  21. This is a great tool for finding spammers who have yet to be suspended by Twitter.

    My question is, why doesn't twitter have a system in place to automatically unfollow any suspended accounts?

    Or, similarly, is there a tool like this to go through all of your followers and figure out if any of them have been suspended so that you can unfollow them to make room for new followers (if you've reached a twitter limit)?

    love the blog! thanks!!

    http://www.twitter.com/nonpretentious

  22. This is a great tool for finding spammers who have yet to be suspended by Twitter.

    My question is, why doesn't twitter have a system in place to automatically unfollow any suspended accounts?

    Or, similarly, is there a tool like this to go through all of your followers and figure out if any of them have been suspended so that you can unfollow them to make room for new followers (if you've reached a twitter limit)?

    love the blog! thanks!!

    http://www.twitter.com/nonpretentious

  23. This is a great tool for finding spammers who have yet to be suspended by Twitter.

    My question is, why doesn't twitter have a system in place to automatically unfollow any suspended accounts?

    Or, similarly, is there a tool like this to go through all of your followers and figure out if any of them have been suspended so that you can unfollow them to make room for new followers (if you've reached a twitter limit)?

    love the blog! thanks!!

    http://www.twitter.com/nonpretentious

  24. This is a great tool for finding spammers who have yet to be suspended by Twitter.

    My question is, why doesn't twitter have a system in place to automatically unfollow any suspended accounts?

    Or, similarly, is there a tool like this to go through all of your followers and figure out if any of them have been suspended so that you can unfollow them to make room for new followers (if you've reached a twitter limit)?

    love the blog! thanks!!

    http://www.twitter.com/nonpretentious

  25. You don't -have- to use those terms – those I just illustrated for example. I'd personally go the Google Reader route so that you can use your own judgement to unfollow.

  26. You don't -have- to use those terms – those I just illustrated for example. I'd personally go the Google Reader route so that you can use your own judgement to unfollow.

  27. You don't -have- to use those terms – those I just illustrated for example. I'd personally go the Google Reader route so that you can use your own judgement to unfollow.

  28. You don't -have- to use those terms – those I just illustrated for example. I'd personally go the Google Reader route so that you can use your own judgement to unfollow.

  29. That's trickier – check out the API wiki – linked at the bottom of every Twitter page.

  30. That's trickier – check out the API wiki – linked at the bottom of every Twitter page.

  31. That's trickier – check out the API wiki – linked at the bottom of every Twitter page.

  32. That's trickier – check out the API wiki – linked at the bottom of every Twitter page.

  33. I never use Yahoo Pipes, this gives me a great excuse. Thanks Chris-this was a great post.

  34. I never use Yahoo Pipes, this gives me a great excuse. Thanks Chris-this was a great post.

  35. I never use Yahoo Pipes, this gives me a great excuse. Thanks Chris-this was a great post.

  36. I never use Yahoo Pipes, this gives me a great excuse. Thanks Chris-this was a great post.

  37. This is a great tool for finding spammers who have yet to be suspended by Twitter.

    My question is, why doesn’t twitter have a system in place to automatically unfollow any suspended accounts?

    Or, similarly, is there a tool like this to go through all of your followers and figure out if any of them have been suspended so that you can unfollow them to make room for new followers (if you’ve reached a twitter limit)?

    love the blog! thanks!!

    http://www.twitter.com/nonpretentious

  38. You don’t -have- to use those terms – those I just illustrated for example. I’d personally go the Google Reader route so that you can use your own judgement to unfollow.

  39. That’s trickier – check out the API wiki – linked at the bottom of every Twitter page.

  40. This is a great tool for finding spammers who have yet to be suspended by Twitter.

    My question is, why doesn't twitter have a system in place to automatically unfollow any suspended accounts?

    Or, similarly, is there a tool like this to go through all of your followers and figure out if any of them have been suspended so that you can unfollow them to make room for new followers (if you've reached a twitter limit)?

    love the blog! thanks!!

    http://www.twitter.com/nonpretentious

  41. You don't -have- to use those terms – those I just illustrated for example. I'd personally go the Google Reader route so that you can use your own judgement to unfollow.

  42. That's trickier – check out the API wiki – linked at the bottom of every Twitter page.

  43. I never use Yahoo Pipes, this gives me a great excuse. Thanks Chris-this was a great post.

  44. I never use Yahoo Pipes, this gives me a great excuse. Thanks Chris-this was a great post.

Leave a Reply

Your email address will not be published. Required fields are marked *

Pin It on Pinterest

Shares
Share This