How does a spam filter work?

Nowadays, almost all email service providers can automatically detect spam emails in user accounts effectively and redirect those potential spam emails to spam folders without human intervention. But how are spam emails detected automatically by email service providers?

Almost all email service providers use machine learning to detect these spam emails. Typically, this machine learning technique relies on some predefined rules. When an incoming email matches most of those rules, the email is marked as spam and redirected to spam folders automatically. Otherwise, the email is sent to the inbox. (What is a Bayesian spam filter?)

What is a spamtrap?

To detect spam emails automatically, firstly, one has to decide on rules for detecting spam emails, based upon which the software can detect potential spam emails. To decide on those rules, enough research is done on spam emails to detect the most common properties of spam emails. Based on those properties, rules for detecting spam emails are set. Once the rules are decided, the email service providers set those rules in the spam detection software. And, spam emails are automatically detected in user email accounts.

A spamtrap is an email address that is used to collect spam emails so that enough research can be done on them to detect spam.

We have learned about honeypots and how they are used to lure the attackers. Spamtraps are like honeypots (What is a honeypot?) for collecting spam emails. They are the email addresses that are meant to collect spam emails only.

How are spamtraps used?

Anti-spam systems are usually automated. They collect samples of spam emails and make rules based on them. Spamtraps, which are email addresses dedicated to receiving spam emails only, are first created. They collect spam emails. After enough sample spam emails are collected, the anti-spam system studies them and makes rules for detecting spam. And everything is done in an automated way.

How do spamtraps reach the spammers?

After creating spamtraps they are published over the Internet. When spammers collect email addresses from various websites using crawlers, the spamtraps are collected by the crawlers. As anti-spam systems work in an automated fashion, any legitimate emails coming in the spamtraps can be mistakenly taken as spam …