Google’s John Mueller answered a question about how Google’s algorithm handles abbreviations. John answered the question in depth, explaining that it’s basically synonyms and that Google doesn’t do anything particularly special with abbreviations.
How does Google handle abbreviations?
The questioner wanted to know how Google handles abbreviations like “for example” which stands for ergo.
They said they had a lot of these kinds of abbreviations on their site.
Google’s John Mueller replied:
“And the short answer is that we don’t do anything special with this stuff.
We essentially treat them like tokens on a page.
And a token is basically a bit like a word or phrase on a page.
And we would probably recognize that there are known synonyms for some of them and figure that out a bit.
But we wouldn’t do anything specific here as we would like to have a glossary of what this abbreviation means and handle that in a specific way.
So it’s something where, especially when it comes to synonyms, our systems learn them over time.
And for the most part, we process them when people search, not when we index. »
John Mueller then recommends watching a video of Google search engineer Paul Haahr speaking at a search conference published by Google Search central in which he explains how search works, in which Paul discusses the using synonyms for query expansion.
Muller says:
“And in this video, he goes over some of the synonymous challenges that we’ve had in the past.
And I found that super interesting to watch and it probably also gives you some ideas of how we might handle some of these types of expansions as far as abbreviations go.
Mueller said the video was from December 2019 or 2020 and was posted on Google Search Central’s YouTube channel.
And there is indeed a video from 2019 that was posted in 2020 in which Paul Haahr talks about synonyms and expanding queries.
Paul touches on the subject of expanding queries and synonyms at minute 1:30:
Google video of Paul Haahr discussing synonyms and expanding queries
“So first I’m going to talk about something in one of our language understanding systems, which is the synonym system.”
The screen behind Paul displays the following text:
-
-
"User vocabulary ≠ Document vocabulary
-
System tries to bridge the gap by automatically adding alternative words
-
Similar to using OR, but usually less important than original terms
-
One of Google Search's most important components"
-
Paul Haahr explains the purpose of using synonyms:
“So what is our system of synonyms?
It’s something that’s there to bridge the gap between the user’s vocabulary and the query’s vocabulary – or the user’s vocabulary and the document’s vocabulary.
In other words, when we see a request, it is often written in a different language than the documents used.
And we try to match those things.
The way it actually works is that it looks like we’re adding a bunch of terms with OR.
Who here in the audience has used the OR operator?
And the way our synonym system works efficiently, we take a user query and add a lot of OR terms to it.
And it’s actually one of the most important ranking things in Google…it’s sort of…it’s something that we started about 15+ years ago that has improved a lot over the years.
More videos of Paul Haahr discussing synonyms and expanding queries
There’s another video of Paul Haahr speaking at SMX West in 2016 where he also talks about query expansion and that sheds some light on the subject as well.
Paul discusses the extension of requests at minute 6:35 of the presentation:
Watch Google Engineer Paul Haahr Discuss Synonyms
This is not the video that John Mueller was referring to, but it also contains some interesting information.
Paul Haahr explains:
“We do a query understanding part where we try to figure out what the query means, we do some fetching and scoring…and then we make some…tweaks.
So Query Understanding, the first question is, do we know any named entities in the query?
The San Jose Convention Center, we know what it is. Matt Cutts, we know what it is.
And so we label them.
And then, are there any useful synonyms?
Does General Motors in this context… does GM mean General Motors?
Does GM mean genetically modified?
And what I mean is that context matters.
We examine the entire query for context. »
Google and abbreviations
What Mueller seems to be saying in his answer is that Google treats abbreviations as synonyms. So when thinking about how Google might understand a content page, an abbreviation can be condensed into a basic meaning that could be considered a synonym.
Quote
How Google handles abbreviations
Watch John Mueller answer the question at minute 48:43
Comments are closed.