Google takes the use of Tensorflow, a home-grown and open-source machine learning (ML) framework to fix Gmail’s Spam problem. Gmail hitherto features traditionally standard anti-spam system, which fails to counter the issue effectively. So, it seems the Silicon Valley giant has come forth with advanced built-in security protections.
ALSO READ: Google’s Instant Tethering now available on more devices
The world’s leading free email service boasts a monthly userbase around 1.5 billion. Another impressive figure would be its 5 million paying business users who use it in the workplace.
How does Gmail Block Spam Mails?
The use of TensorFlow AI framework compliments existing ML and rules-based protections, which is likely to improve the spam-detection detection capabilities. Google claims that it could now block a whopping 100 million additional spam messages every day.
Spam categories that were earlier difficult to detect would be blocked henceforth. TensorFlow is used to block image-based messages, emails with hidden embedded content, and messages from newly created domains that try to conceal a low volume of spammy messages within legitimate traffic.
ALSO READ: Facebook rolls out ‘Unsend’ message feature in the Latest Update
Google’s AI learns patterns based on user’s mail consumption, behavior, and decisions. So, the intent is not to block messages barely on the email characteristics that match up to those commonly considered “spammy.” Instead, the new mechanism would personalize the protection for each user. Finally, the AI would timely self-enhance the model and refine the granular decisions to effectively tackle spam.
