Gmail spam catching prowess improved with neural network

Shane McGlaun - Jul 10, 2015, 7:00 am CDT
0
Gmail spam catching prowess improved with neural network

Spam email is one of the most annoying things about the internet today with inboxes flooded with messages for cheap medications and dating sites that few people actually want to see. Gmail is pretty good at catching spam messages and hiding them away in the spam folder without most users having to do much to make it happen. However, we all know that at some point you need to check spam because inevitably a message that isn’t spam ends up in there.

Google has announced some improvements to Gmail that have to do with detecting spam and reducing misfiled messages. One of the ways it is improving spam detection is by helping legitimate senders of lots of email get their messages delivered via Gmail Postmaster Tools. These tools help big email senders to analyze email and gives them data on delivery errors, spam reports, and reputation so they can tell if they are likely to run into issues.

The big change for Gmail is that the spam filter now uses an artificial neural network to detect and block sneaky spam messaged. The new neural network allows the spam filters to be customized for individual user tastes. That means if you get a lot of newsletters that you want to read via email, they won’t get field in spam and if your neighbor hates newsletters, he can send them all directly to spam.

Google also notes that the spam filter is now better than ever at rooting out email impersonation, which is where many phishing attacks come from. New machine learning signals allow Gmail to determine if an email actually came from the sender it says it did.

SOURCE: Gmail


Must Read Bits & Bytes