A taste of tweets: reverse engineering Twitter spammers

Abstract

In this paper, through reverse engineering Twitter spammers' tastes (their preferred targets to spam), we aim at providing guidelines for building more effective social honeypots, and generating new insights to defend against social spammers. Specifically, we first perform a measurement study by deploying "benchmark" social honeypots on Twitter with diverse and fine-grained social behavior patterns to trap spammers. After five months' data collection, we make a deep analysis on how Twitter spammers find their targets. Based on the analysis, we evaluate our new guidelines for building effective social honeypots by implementing "advanced" honeypots. Particularly, within the same time period, using those advanced honeypots can trap spammers around 26 times faster than using "traditional" honeypots. In the second part of our study, we investigate new active collection approaches to complement the fundamentally passive procedure of using honeypots to slowly attract spammers. Our goal is that, given limited resources/time, instead of blindly crawling all possible (or randomly sampling) Twitter accounts at the first place (for later spammer analysis), we need a lightweight strategy to prioritize the active crawling/sampling of more likely spam accounts from the huge Twittersphere. Applying what we have learned about the tastes of spammers, we design two new, active and guided sampling approaches for collecting most likely spammer accounts during the crawling. According to our evaluation, our strategies could efficiently crawl/sample over 17,000 spam accounts within a short time with a considerably high "Hit Ratio", i.e., collecting 6 correct spam accounts in every 10 sampled accounts.