from the what-you-can-learn-from-reading-spam-headlines dept
Oh, those nutty spammers. Someone should do a study of spam headlines to see if they’re a good way to tell what subject is popular in the news at the time the spam was sent. We’ve already written about how quickly spammers like to pick up on current events, and now comes news that many spammers are mentioning the high price of gas in the subject lines of their emails to get people to click. Apparently 5% of spam mentioned gas in the subject line one day last month. Some of the gasoline spam is actually related to gasoline – but is really trying to get people to sign up to get spammed some more. Plenty of other spam just uses gas in the subject line to get you to click – at which point you’ll be presented with porn. So, maybe you shouldn’t look for “Sexually-Explicit” in the subject line to identify your porn spam – but, rather “Gasoline.”
Comments on “Gasoline Spam”
Someone should do a study of spam headlines to see if they’re a good way to tell what subject is popular in the news at the time the spam was sent.
Unfortunately don’t think I have the data or code around any longer, but I did exactly this a while ago. Frequency analysis of the words appearing in the spam that I received (this was back before spam message randomizing was such a big thing, so I included the content as well as subject) and the top news stories of the day scraped from a few different Web sites.
Basically what I figured out what that the overwhelming majority of spam doesn’t appear to be influenced by current events; there definitely is some “up to the minute” spam out there, but I would have needed a much bigger spam corpus if I was going to be able to learn much of anything about that topical spam. This was back in the old days, after all, when my filters were only catching about a hundred or so spam messages a day… 🙂