Subscribe

Gmail prevents 'sneaky spam'

Lauren Kate Rawlins
By Lauren Kate Rawlins, ITWeb digital and innovation contributor.
Johannesburg, 13 Jul 2015
Google improves Gmail spam filters for the user, while allowing bulk-senders to track e-mail performance.
Google improves Gmail spam filters for the user, while allowing bulk-senders to track e-mail performance.

Google has added new technology to its e-mail service to guarantee spam does not reach users' inboxes, but ensures legitimate e-mail does.

E-mail marketers and bulk senders can now make use of Postmaster Tools to analyse and track e-mail performance.

Less than 0.1% of e-mail in the average Gmail inbox is spam, and the amount of wanted mail landing in the spam folder is under 0.05%, says Sri Harsha Somanchi, Google product manager, in a blog post.

Messages such as monthly bank statements and airline ticket receipts are sometimes mistakenly classified as spam, explains Somanchi. The new Gmail feature will prevent this happening by conveying to the sender that the e-mail did not make it to a primary inbox.

The new Gmail feature will give senders data on spam reports, delivery errors and reputation. "This way they can diagnose any hiccups, study best practices, and help Gmail route their messages to the right place."

Somanchi says up until now, the Gmail filter learnt from end-users by allowing them to 'report spam' in their regular inboxes and press a 'not spam' button in their spam inbox.

"Now, we are bringing the same intelligence developed for Google Search and Google Now to make the spam filter smarter in a number of ways. For starters, the spam filter now uses an artificial neural network to detect and block the especially sneaky spam ? the kind that could actually pass for wanted mail."

Machine learning will also play an important role in curating e-mail preferences, meaning spam filters will work differently based on what individual users mark as spam.

"Finally, the spam filter is better than ever at rooting out e-mail impersonation ? that nasty source of most phishing scams. Thanks to new machine learning signals, Gmail can now figure out whether a message actually came from its sender, and keep bogus e-mail at bay," concludes Somanchi.

Share