On average, less than 0.1 percent of email in the Gmail Inbox is spam, Google says. The average percentage of important mail being rerouted to the spam folder, on the other hand, is even lower, at about 0.05 percent.
Even so, Google wants to enhance Gmail's spam filter further to give its users "the mail you want, not the spam you don't." The company says it is finding new ways to support senders of important mail while also developing the latest in spam filtering.
This type of support applies to email messages such as a bank's monthly statement, an airline's travel itinerary receipt or any of those messages that the user expects to see in their inbox in a timely manner.
There are times, however, when these messages, otherwise called "wanted message," land in the spam folder by mistake. Users, of course, check their inbox and not so much their spam folder. When the user does open the spam folder, it is usually done out of curiosity.
In order to help authentic email senders get their message across, Google created Postmaster Tools, which use a smart algorithm in analyzing emails from high-volume senders. It also performs other unique services, such as spam reports and reputation and data checks on delivery errors.
According to Google, Gmail's new spam filter uses the same intelligence that is already being used in Google Now and Google Search. Dubbed an artificial neural network, the spam filter is enhanced to detect and block sneaky spam.
"Sneaky spam" refers to the type of mail that has landed in one's inbox "successfully" even if it's supposed to be classified as spam.
The new spam filter is so smart that it can actually reflect individual preferences.
"We also recognize that not all inboxes are alike," wrote Sri Harsha Somanchi, product manager at Gmail. "So while your neighbor may love weekly email newsletters, you may loathe them."
Lastly, Gmail can now detect whether the message was really sent by its true sender and not by an impersonator who is phishing for information.
Together with Twitter and Facebook, Google is one of the leaders in neural networking and has successfully applied the technology to a number of its services. In Twitter's case, the company plans to use the technology to detect not only junk mail but also unwanted hate messages.
"We're starting to use it for more textual understanding, so we can identify spam and abuse," said Alex Roetter, Twitter's head of engineering.
Google will continue enhancing the Gmail experience for its users, ensuring that their inboxes are free from spam and phishing.
Photo: Cairo | Flickr
