Google highlights new machine learning features heading to Chrome
Google uses machine learning (ML) models to offer a host of useful features on almost all of its products, and the company’s popular browser app, Google Chrome, is no exception. Chrome already offers several ML-powered features across all platforms. These include features to make web images more accessible for the visually impaired and real-time captions for videos to help those hard of hearing. But these aren’t the only ML features found on Google Chrome.
In a recent blog post, Google highlighted some recently released ML features that are now making their way to more Chrome users. Additionally, the company has revealed a few other new features that should reach users with future builds.
Safe Browsing in Google Chrome, for instance, is an ML-powered feature that shows warnings when users try to navigate to dangerous websites or download malicious files. It has been around for a while, but Google recently rolled out a new ML model that identifies 2.5 times more potentially malicious sites and phishing attacks than the previous model, making it an even more invaluable tool for Chrome users. The feature can also silence potentially malicious notifications from websites, and Chrome will soon be able to do all this entirely on-device.
To that end, Google plans to roll out another new ML model with the Chrome M102 release. It will help the browser make the aforementioned notification predictions entirely on-device. As the company notes:
“To further improve the browsing experience, we’re also evolving how people interact with web notifications. On the one hand, page notifications help deliver updates from sites you care about; on the other hand, notification permission prompts can become a nuisance. To help people browse the web with minimal interruption, Chrome predicts when permission prompts are unlikely to be granted, and silences these prompts. In the next release of Chrome, we’re launching an ML model that is making these predictions entirely on-device.”
Furthermore, Google is using machine learning to improve the Journeys feature that rolled out earlier this year. In case you’re not familiar, the Journeys feature in Google Chrome organizes your browsing history into categories with (what Chrome thinks are) the most important pages in a given category/topic.
Google now plans to use ML to make these categorized websites available in your preferred language as soon as you return to them. The company has launched an updated language identification model “to figure out the language of the page, and whether it needs to be translated to match your preference.”
Lastly, Google plans to use ML to adjust Chrome’s toolbar in real-time to highlight the most useful actions for you at any given moment. As shown in the attached screenshots, the button next to the omnibar will change automatically, based on what Chrome things you wish to do next — share the webpage or perform a voice search. In case you’re wondering, Chrome will still give you an option to customize the toolbar as per your needs when the feature rolls out with a future build.
What do you think of the new machine learning features coming to Google Chrome? Let us know in the comments section below.