Spammers Are Becoming "Smarter" on Twitter

T witter has become one of the most commonly used communication tools in daily life. With 500 million users, Twitter now generates more than 500 million tweets per day. However, its popularity has also attracted spamming. Spammers spread many intensive tweets, which can lure legitimate users to commercial or malicious sites containing malware downloads, phishing, drug sales, scams, and more.1 Spam is a problem throughout the Internet, and Twitter is not immune. In addition, Twitter spam is much more successful compared to email spam.2 Various methods have been proposed by researchers to deal with Twitter spam, such as identifying spammers based on tweeting history3 or social attributes,4 detecting abnormal behavior,5 and classifying tweet-embedded URLs.6 Although researchers, as well as Twitter itself, have attempted to combat spam, the percentage of spam in the whole platform is still high. We hypothesize that this is because spammers are becoming more cunning on Twitter. While researchers are developing methods to detect spam, spammers continuously invent new strategies to bypass detection. Here, we review the well-known methods that spammers have used to avoid or reduce their chances of being caught on Twitter. We show that spammers are now using more advanced strategies, namely coordinated posting behavior, finitestate-machine-based spam template, and passive spam.