Google Chrome has received a new update and among its improvements, we have found one that may be of great interest to users of this browser. To be exact, the main improvement brought by this update (M124) is the one that affects its address bar, called Omnibox by the Mountain View company, and therefore, to all of us who use it in our daily lives.
The way this is going to experience a substantial improvement is by making its users receive better recommendations through machine learning. In this way, we should start receiving more appropriate recommendations for what we need and search for. This novelty replaces the algorithms that the mentioned Chrome address bar used until now to offer its recommendations.
Google Chrome’s address bar suggestions are about to change forever
The previous operation of this Omnibox to display its recommendations was to use algorithms, which assigned scores to each recommendation, but this method is outdated for modern needs, according to the developers themselves. Therefore, this is the reason why the recommendations themselves have not changed for quite some time and what has led to the implementation of these machine learning models.
However, the scoring system remains, but with this new update, the score is reduced or increased based on the user’s actions, which will make Omnibox offer us more accurate and appropriate recommendations. In other words, now Google can “collect more recent signals, retrain, evaluate, and implement new models periodically over time,” as they indicate in their statement on the official Chromium blog.
The operation of this new method can be explained simply with this example: if we type in the address bar and enter a recommended address, but then immediately click on the bar again, Chrome will be aware that this page was not the one we were looking for and will lower its score. Therefore, that suggestion will appear further down in the recommendations, reaching a point where it may not even appear if its score is very low.
They have also talked about the future and the potential that this new method of machine learning has. For example, Chrome developers indicate that, possibly, in the future they will be able to “potentially incorporate new signals, such as differentiating between the time of day to improve relevance” so that we can have even more accurate recommendations. In addition, it is also indicated that they want to “explore training specialized versions of the model for particular environments: for example, mobile, business or academic users, or perhaps different locations”.
This new machine learning system for the Google Chrome address bar should already be active, as long as you have the latest version of the browser from Mountain View. If you don’t see any changes, we assume it’s just a matter of time before you start noticing the changes in your recommendations.
