Try for Free
Advertisement
  1. SEJ
  2.  » 
  3. News

Google Begins Using BERT to Generate Top Stories Carousels in Search

Google Begins Using BERT to Generate Top Stories Carousels in Search

Google is now using BERT models, along with other machine learning techniques, to group related news articles together in carousels.

The new BERT-generated carousels will appear in the top stories section of Google Search. They’re designed to help people easily find the best articles related to a particular news story.

Related: BERT Explained: What You Need to Know About Google’s New Algorithm

Google’s Duncan Osborn, product manager of Search, explains in an announcement:

“When you’re searching for information on a timely topic–a recent sports upset or the latest scientific breakthrough–you see a carousel of articles at the top of your Search results highlighting relevant news. Now, when there are multiple stories related to your search, we’ll also organize the results by story so it’s easier to understand what’s most relevant and you can make a more informed decision on which specific articles to explore.”

In addition to surfacing the most recent coverage, Google’s new top stories carousels will surface other content that will help provide more context and perspective to a given story.

For example, searches for “NASA news,” might return results grouped under distinct news stories – such as “NASA adds five companies to moon bid” and “NASA detects water vapor on Jupiter’s moon Europa,” – along with additional results under “Also in the news.”

Google Begins Using BERT to Generate Top Stories Carousels in Search

Google Begins Using BERT to Generate Top Stories Carousels in Search

Google further explains how machine learning is used to generate these results:

Advertisement
Continue Reading Below

“To power this new experience, we developed a new story-understanding technology to map the people, places and things involved in a news story, and then draw connections between them. This technology allows us to create distinct groups of news articles. To generate these groups, we use a variety of machine learning techniques including BERT models to examine the related articles and determine where one story ends and another begins.”

According to Google’s research, clustering results into clearly-defined stories is beneficial in helping people navigate search results for news stories and find the most relevant content.

Lastly, top stories carousels will now contain ancillary information such as notable quotes and related opting on pieces. Google notes these changes are part of a long-term effort to expand the types of news coverage it shows users in search results.

Advertisement
Continue Reading Below

These updates are available starting today on mobile devices in US English search results, and rolling out to more languages and locations in the coming months.

Related:

ADVERTISEMENT

Subscribe to SEJ

Get our daily newsletter from SEJ's Founder Loren Baker about the latest news in the industry!

Ebook

Matt Southern

Lead News Writer at Search Engine Journal

Matt Southern has been the lead news writer at Search Engine Journal since 2013. With a degree in communications, Matt ... [Read full bio]

ADVERTISEMENT
Advertisement
Read the Next Article
Read the Next