Chrome: this is how Google changes its address bar to improve the relevance of its browser

Google announces changes to Omnibox, its browser’s search bar Chromium. It should be optimized using machine learning.

When we browse Chrome, the Omnibox search bar allows us to return to previous searches, which we can do regularly, or to suggestions from the browser, which is based on the first letters typed. . A fairly convenient service, which should be even more relevant in the future thanks to the integration of machine learning models into this search bar.

From hand-tuned formulas to machine learning

Omnibox is seeing its mode of operation change profoundly. While the prediction system of this search bar previously worked thanks to “ formulas built and adjusted by hand », it will now benefit from machine learning.

The goal here is to make the Omnibox more flexible and capable of “ collect fresher signals, retrain, evaluate and deploy new models », in the words of the Mountain View firm. The quality of the bar’s proposals should therefore improve over time.

A change that starts with Chrome 124

And to do this, the Omnibox will take into account the behavior of the Internet user. If the search bar, after suggesting a URL, records that the user only stays a short time on the proposed page, it will give this page a low rating, meaning that it was not the right result to offer .

For Google, machine learning in the Omnibox represents a valuable tool for the future. He would open “ many new opportunities to improve the user experience by potentially incorporating new signals, such as differentiating times of day to improve relevance. » And this change is integrated into the latest Chrome 124 update. Will we see anything new in a few weeks?

Source : Chromium, Neowin