Google Announces BERT, its Largest Algorithm Update to Date

Google is getting smarter, and it’s driven by advances in natural language processing.

Today, Google announced an update to search, made possible by state-of-the-art natural language processing advances. Particularly in the pre-training of contextual representations.

“BERT AI is special, because it is the first deeply bidirectional, unsupervised language representation, plainly trained using only a plain text corpus.” – Source

What Google has essentially done, is apply BERT to another text corpus – the content it scraped from your web pages.

This means that Google is now able to better match your query to the content you wrote. Not just a little better – much better.

Here are some examples from Google:

Do Estheticians Stand a Lot at Work — ” Previously, our systems were taking an approach of matching keywords, matching the term “stand-alone” in the result with the word “stand” in the query. But that isn’t the right use of the word “stand” in context. Our BERT models, on the other hand, understand that “stand” is related to the concept of the physical demands of a job, and displays a more useful response.” – Google

Parking On a Hill With No Curb — ” In the past, a query like this would confuse our systems–we placed too much importance on the word “curb” and ignored the word “no”, not understanding how critical that word was to appropriately responding to this query. So we’d return results for parking on a hill with a curb!” – Google

Math Practice Books For Adults — ” While the previous results page included a book in the “Young Adult” category, BERT can better understand that “adult” is being matched out of context, and pick out a more helpful result.” – Google

Can You get Medicine For Someone Pharmacy — “With the BERT model, we can better understand that “for someone” is an important part of this query, whereas previously we missed the meaning, with general results about filling prescriptions.” – Google

Three years ago, I began working on the Rank Candidate Theory, which basically asserts that Google has been doing exactly this kind of processing for a while now, and how it relates to how Google ranks content.

The AI that we ended up building is designed to look at content Google is favoring in search with the goal of understanding WHY. As Google results are getting smarter, the INK editor is able to take advantage of these refinements and will better understand what BERT is looking for.

INK is free for all to download and use. Our latest update, released today, will help you to check how your existing content is affected by Google BERT.

Google Announces BERT, its Largest Algorithm Update to Date

Found this article interesting?

Alexander De Ridder

Comments (0)

249

Link Copied Successfully

Google Announces BERT, its Largest Algorithm Update to Date

Found this article interesting?

Alexander De Ridder

Comments (0)

249

Link Copied Successfully

Sign in