The Mountain View-based tech giant recently published an official blog post explaining how the search predictions in Google Search work. In this post, the company revealed how Autocomplete’s predictions are generated using recent trends, locations, and languages. “To determine what predictions to show, our systems begin by looking at common and trending queries that match what someone starts to enter into the search box,” reads the post. So, whenever you start typing in the Google search box, the engine immediately starts looking for related searches in the background. However, that is not all.
There’s More to It…
“We don’t just show the most common predictions overall. We also consider things like the language of the searcher or where they are searching from because these make predictions far more relevant,” adds the post. As an instance, the company compared (picture) the search predictions of “driving test” for a resident of the US state of California with the search predictions of the same for a resident of the Canadian Province of Ontario.
So, as you can see, the predictions for the residents of different locations differ hugely. The company also notes that as the engine considers the language of the searcher, the spelling for “center” is “centre” for the Canadian resident. Apart from the above considerations, Google mentions that they “also take freshness into account when displaying predictions”. This means that if there is a trending topic on the internet, the search engine might show a “trending prediction even if it isn’t typically the most common of all related predictions that we know about.” Now, the company states that although these search predictions are meant to help users search for their queries faster, sometimes they might also show irrelevant or shocking results. “We deal with these potential issues in two ways. First and foremost, we have systems designed to prevent potentially unhelpful and policy-violating predictions from appearing. Secondly, if our automated systems don’t catch predictions that violate our policies, we have enforcement teams that remove predictions in accordance with those policies,”adds the post further.
