In a blog post yesterday, Google announced, that its spam filters have become more intelligent than before. Google had relied on Machine learning (like most mailing services out there) till now for checking spams. However, its latest approach involves an artificial neural network, the same tech which the company deploys behind Google Now and Google Search.
Before we get on to how Google improved its spam filters, lets get to know very briefly on how the previous machine learning tech worked. Currently, when you click the “Report spam” and “Not spam” buttons, you’re not only improving your own Gmail experience right then and there, you’re also training Gmail’s filters to identify spam vs. wanted mail in the future.
Now, Google’s artificial neural network, the tech behind its powerful Google Search and Google Now tools, learns from your feedback to detect and block the especially sneaky spam—the kind that could actually pass for wanted mail.
Another major learning curve, which Google’s artificial brain learned from user’s spam reporting feedback, is that every user has his/her own needs of newsletters. While some like a daily update, the more relaxing ones want a weekly one. With advances in machine learning, the spam filter can now reflect these individual preferences.
Most importantly though, Google says that Gmail spam filters are now better than ever at rooting out email impersonation—that nasty source of most phishing scams. With the help of new machine learning signals, Gmail can now figure out whether a message actually came from its sender, and keep bogus email at bay.
So you are a bulk sender, a newsletter sender who happens to be on the others side of affairs ?
Well, more stringent spam filters might well cut down your delivery and conversion rates, but Google has separately introduced something called Postmaster Tools, which, as the name implies, will let a select group of “qualified high-volume senders” send bulk mails to gmail users. Whether you get into that list or not is solely on the discretion of Postmaster Tools’ own machine learning algorithm. Google though, on its part, hasn’t really mentioned as to what factors it will be taking into account to select its “qualified high-volume senders”.