Who doesn’t hate email spam clogging up their inbox daily? Google is now working to fix this, with the Gmail team and Google recruiting its in-house machine learning framework, TensorFlow, to help train additional spam filters for Gmail users.
Google has said, via the Gmail Blog overnight, that Gmail is already blocking an extra 100 million spam messages every day. To put it into context, Gmail already blocks 99.99% of spam for its 1 billion plus users, meaning there is still 0.1% of spam that still comes through to users inbox.
This is where TensorFlow will help catch the spammers who slip through, without accidentally blocking messages that are important to users.
Neil Kumaran, product manager of Counter Abuse Technology at Google has told The Verge that:
“At the scale we’re operating at, an additional 100 million is not easy to come by,” and that “Getting the last bit of incremental spam is increasingly hard, [but] TensorFlow has been great for closing that gap.”
TensorFlow has been added alongside existing Artificial Intelligence (AI) and rule-based filters that Gmail has utilised for years. Though it is important to point out that rule-based filters can can block the most obvious spam, machine learning looks for new patterns spammers maybe using to help suggest an email is not to be trusted.
By using TensorFlow’s AI technology, it make managing data at this scale easier, while the open-source nature of framework means new research from the community can be quickly integrated.
Google launched TensorFlow back in 2015 and it has very quickly become an incredibly important part of its AI business. It has been praised both for its flexibility and capacity to scale alongside working with Google’s other AI services.
Google has said that, by integrating TensorFlow into Gmail, it can better personalise spam filters and that this is the turning point for understanding spammers signals and “turning those signals into better results.”