Any Twitter user knows that, Spam is a problem and that the social network has its fair share of it. A team of researchers is trying its best to fight against such silly people. Reports from a paper released this week claim that, they detailed how they’ve been working with Twitter to take a close look at how fraud and fake accounts are made and what should be done in order to avoid them.
Success seems to be close as, a team of researchers from George Mason University, which is the International Computer Science Institute, and the University of California, Berkeley, have together worked with Twitter to purchase over 127,000 fraud accounts which have been generated automatically from 27 different merchants in a period of 10-months starting in June 2012.
Their goal is to try and develop a simple way to stop spam accounts before they’re created or before they’re used as a wrong source to spread attempts, scams, malware, phishing, and more across the web. As per the results, it seems that the team has been successful and with the help of Twitter’s assistance, they were able to work up on a set of identifiers which could be used to sniff out when accounts are automatically generated. The researchers were able to flag usernames specifically made using specific patterns, and used information on the signup procedure.
Twitter has ultimately used the data to wipe out “several million” of the fraud and fake accounts which came from the 27 different merchants which the researchers studied, and has been helpful in catching new accounts as well. As Twitter implemented the changes, 90 percent were dead on arrival, when the researchers attempted to buy 14,000 additional accounts. One of the merchants told the researchers that, not only his but all of the stock had been suspended and exclaimed that he had no idea as to what Twitter did.
Twitter spam is gone for good is not what all this means. As per the paper it’s estimated that the accounts stemming from these merchants represent about 10 to 20 percent of total Twitter spam. One of the researchers told Brian Krebs for his security blog that, they would love to keep doing this, but the hard part they face is they kind of have to keep doing the buys which accounts for lot of work as it has to be updated.
54 percent of new accounts purchased by the researchers were immediately suspended as just two weeks after Twitter started using the team’s work. Although, the team made a few recommendations from their research, there are many other ways to counteract spam which currently costs just about four cents per Twitter account. But still, Twitter could stop spammers with higher costs by making it more difficult to generate fake accounts.
Twitter is in the need of an email address to sign up, but researchers found that merchants often didn’t pass along the email credentials used to make spammy accounts, meaning periodic requests for email confirmation after signup could catch some spammers. According to the research, Gmail accounts cost up to 150 times more than their Hotmail and Yahoo equivalents, as Google requires phone verification to sign up.
It is quite unsurprising as over 60 percent of the fraud accounts studied were connected to Hotmail and around 11 percent with Yahoo, whereas only 1.89 percent used Gmail. Although the researchers claim that the codes had successfully stopped 92 percent of accounts from being made, only 35 percent of accounts the researchers purchased required a CAPTCHA. However with more adaptive IP blacklists or stricter CAPTCHA requirements, Twitter can possibly stem the rising tide against spammers.