Matt vs Spam: My No Spam Framework

Today, I remembered that I started an anti-spam framework project a while back. I’d done a lot of the legwork and just got distracted at some point. So, I put in some hours and progressed the project. It works using some pretty old-school anti-spam techniques as well as a few of my own invention. The framework loops through a collection of different tests and gives the comment or signup in question a score to determine how spammy it looks. I started the project because a small handful of simple patterns could remove most of my spam from ever seeing my moderation queue. When it is tested and I am ready to share, it should have a nice collection of rules, a way to grab an IP address list of shameful spammy domains, a Bayesian filter so you can train it to get better at classifying spam, and – because it is now 2026 – it should be possible to ask an AI for its opinion too. Those last two are important because spam is evolving and uses generative AI quite effectively. Follow me for…