In the list of Al uses, Spam blocking is another thing that you can perform with it. Google yesterday announced that they are now using AI specialized TensorFlow to help train additional spam filters for Gmail users. With the help of this new technique, the company claims to now blocking an extra 100 million spam messages every day.
“At the scale, we’re operating at an additional 100 million is not easy to come by,” Neil Kumaran, product manager of Counter Abuse Technology at Google, tellsThe Verge. “Getting the last bit of incremental spam is increasingly hard, [but] TensorFlow has been great for closing that gap.”
Gmail has one billion plus users as TensorFlow is not a huge gain for the company, it makes one extra blocked spam email per 10 users. But the company said Gmail already blocks 99.9 percent of spam. For years, Gmail is using AI in addition to rule-based filters, it works by blocking the most obvious Spam messages. Machine learning Framework always looks for new patterns that might suggest an email is not to be trusted.
Algorithms balance a huge number of metrics from formatting an Email to the time of the day it has been sent. TensorFlow makes managing the data easier whereas the open-source nature of the framework means new research from the community can be quickly integrated.
In 2015, Google launched TensorFlow and with the passage of time it becomes extremely important for AI business. It is a free machine learning framework which allows developers to create AI tools for a huge range of tasks. It also works effectively with other Google’s AI services which encourage users to buy computing power from Google along with off-the-shelf vision and speech algorithms.
The company said that integrating TensorFlow into Gmail also allow it to better personalize spam filters. This process has taken several years to properly function as Gmail is looking for some signals from which they can judge which email should be spammed.