Google says its Gmail spam filters are now smarter than ever. While the service always used machine learning to figure out what was (and wasn’t) unwanted email, it now uses a more sophisticated neural networking approach to analyze incoming messages.
The new filters are also better at figuring out your individual preferences, Google says. What’s one person’s spam, after all, could be another’s daily influencer marketing newsletter. Until now, training the Gmail filter by marking messages as spam (or ‘not spam’) mostly trained the overall machine learning algorithm, it seems, but didn’t have a lot of influence on your individual mailbox.
The company says that it has also improved its overall phishing protection “thanks to new machine learning signals.”
For legitimate bulk senders — those who need to send lots of receipts, confirmation emails, etc. — Google today launched its Gmail Postmaster Tools. This new service will allow these qualified high-volume senders to get a better idea of how Gmail treats their emails (it’s basically Google Webmaster Tools for bulk senders).
If you are a spammer or phisher, chances are you can also get some useful info from this tool, though you are probably not the target audience for it (and I’m still going to reply to you about that inheritance that’s waiting for me in Nigeria).
Update: Google tells me the tool is only available to qualified senders who meet its reputation requirement, so spammers are out of luck.
The tool will tell give users access to a number of dashboards, including ones that show how often users report the sender’s emails as spam, for example. Postmaster Tools will also give senders more insight into their domain’s reputation (send a lot of spam and Gmail will downgrade you) and additional info about how Gmail treats their messages.