Technology giant Google has announced that its email service, Gmail, will now be able to detect over 100 million more spam emails every day. Google further said that it will be able to do this because of the recent integration of TensorFlow, which has the ability to catch spammers who may slip through Gmail’s scanning technology. With the implementation of TensorFlow, the popular email service has been able to block messages that were image-based, emails that came with hidden embedded content and those that came from domains that are relatively new.
For those unaware, TensorFlow is an open-source machine learning framework that was developed and Google and it was officially launched in 2015. Over the year, the technology has grown to become a service that is largely popular with developers. According to Google, for Gmail, recognising and blocking spam message has been a very challenging task considering how it has constantly been evolving over the last few years.
Google said that although it can successfully block 99.9% of spam emails with its machine learning, TensorFlow will be able to take care of rest of the 0.1% of it.
Furthermore, since TensorFlow has an open-source nature, any fresh community-based research for fighting spam in a better manner will be implemented very easily. To put it simply, TensorFlow will enhance the machine learning efforts that are being made by Google, so that not as many engineers will be needed to ensure that the spam-protection features of Gmail are working as they well as they should.
Google has said on its blog, “Within Gmail, we’re currently experimenting TensorFlow in other security-related areas, such as phishing and malware, as part of our continuous efforts to keep users safe.”
Speaking about the development to The Verge, Neil Kumaran, product manager of Counter Abuse Technology at Google said, “At the scale, we’re operating at, an additional 100 million is not easy to come by. Getting the last bit of incremental spam is increasingly hard, [but] TensorFlow has been great for closing that gap.”
He also added TensorFlow will personalise spam filters in Gmail and learn from user pattern to provide customised results. This means that what may be spam for you, may not be considered spam for your friend in case they engage with it often.