So Google has one of the best AI technology available on the planet and well they have implemented this on various projects, ranging from self driving cars to mobile phone personal assistants, Google have given its users a taste for there AI technology. Moreover, they have now implemented there very famous AI technology on to another new project which is supposedly a new machine-learning based system which will help in improving the Gmail spam classification capabilities.
Must Read: Samsung M30 specs have been revealed!
Furthermore, Google isnow focused on improving there in-house machine learning framework, which will in turn train new spam filters by which Google will be able to block an extra amount of 100 million spam emails. Now you may think that the figure may sound exaggerated but given the fact the Gmail has over billions of users, yeah not that much now is it? Moreover, Google was already blocking about 99.9 spam emails, which is honestly an amazing feat achieved by the company and now you may think that a mere 0.01 percent may not amount to much but when the final numbers came in, that 0.01 percent amounted to a significantly large number.
Gmail App uses AI to block the flow of spam emails
Moving on, it should also be noted that it becomes increasingly difficult to track spam messages as Google will near the 100 percent mark of blocking these spam emails, this is mainly because the spam messages which do get through have features which may not be detected by these filters. Now the traditional way if finding out whether your message is spam or not requires rule based filters, now here a set of rules are applied on the messages which give out a message score. The score is then used to determine if the message is spam or not. Now the Machine learning models try to detect some more complex patterns where rule based systems may fail. The idea is basically to train the network of huge data which has messages classified as spam and not spam. Google use the Tensorflow to analyze there machine learning algorithms
