Home Latest News Google Uses Artificial Neural Network to Detect Sneaky Spam

Google Uses Artificial Neural Network to Detect Sneaky Spam


Google has shared some details on the systems it has implemented to protect its customers against spam emails. The search giant has also released a new tool to help organizations ensure that their emails don’t end up in the spam folder.

According to Google, the anti-spam features introduced over the past period have led to only less than 0.1 percent of emails in a Gmail inbox being spam. Furthermore, the company says less than 0.05 percent of wanted emails are sent to the spam folder.

In an effort to help companies ensure that the emails they send to their customers don’t end up in the spam folder, Google has launched Gmail Postmaster Tools. The new tool is designed to help companies that send out a high volume of emails conduct an analysis of their messages. Users are provided spam reports, reputation analysis, and data on email delivery errors to help them diagnose potential issues and follow best practices.

While Postmaster Tools is a service for organizations, Google believes it is also beneficial for regular users since they will no longer have to search their spam folder for potentially important information such as bank statements and airline ticket receipts.

In addition to the introduction of Postmaster Tools, Google has revealed a series of steps it has taken to fight spam. The search giant has pointed out that every time users mark an email as spam or not spam, they actually train Gmail’s filters to better distinguish wanted from unwanted emails.

Google says it has applied the intelligence used in Google Search and Google Now to the spam filter.

One of the new spam-fighting techniques involves the use of an artificial neural network for detecting and blocking “especially sneaky” spam.

The spam filter has also been improved to detect email impersonation, which is often used in phishing attacks.

“Thanks to new machine learning signals, Gmail can now figure out whether a message actually came from its sender, and keep bogus email at bay,” explained Sri Harsha Somanchi, a Google product manager.

Machine learning has also been used to teach the spam filter to operate based on the individual preferences of each user. This is important, Google has highlighted, because some users like emails such as newsletters and others don’t.

Recommended for You

Recommended for You

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Close Read More

See Ads