Google announced an improvement to the process for rejecting unwanted requests to show notifications from sites in the Chrome browser. To improve user experience and reduce response times to intrusive subscription offers, the company is updating in-browser machine learning algorithms that automate part of this process.
In the next version of Chrome, the mechanism for predicting the usefulness of such requests for the user will be implemented directly in the browser, without exchanging data with Google servers or any other external resources. The algorithms will identify requests that are highly likely to be rejected by them and block them before they are even displayed on the screen, says SearchEngines.
If it detects phishing attempts, Chrome will show a warning, and if it receives a permission request from a site that is unlikely to be accepted by the user, it will mark this in the browser bar.
NIX Solutions adds that in the future, machine learning technology will make the Chrome browser interface adaptive: the toolbar will adapt in real time to the needs of the user, displaying voice search or “Share” buttons next to the address bar at the moment when the user is most likely to need them.
“Our goal is to create a truly useful browser, and we are excited about the possibilities that machine learning provides for this. In the end, the user experience is the only thing that matters,” the developers say.