Google today announced that it’s super-powering Gmail’s spam detection tech with a bunch of new machine learning technology powered by its own TensorFlow framework. The company says it’s using some of TensorFlow’s capabilities to detect 100 million more spam emails every day.
With Gmail apparently being used by 1.5 billion users every month, the service has a lot of emails to deal with. And although Gmail is decent at detecting basic/slightly-complex spam emails, it often struggles to detect spam emails that are disguised as regular emails. Thanks to TensorFlow, Gmail is now able to block more complex messages, like image-based spam emails, emails with hidden embedded content, and emails from newly created domains that are used to hide low volume of spammy messages within legitimate traffic.
TensorFlow also helps Gmail provided more personalised spam detection — for example, if an email is considered spam by one person, it doesn’t mean it will be considered spam on your inbox, so Gmail make sure to deal with that appropriately.
Google isn’t saying whether TensorFlow will help with the accuracy of spam detection when it comes to flagging non-spam email as spam, but the personalised spam detection should help.