How Amazon Comprehend Powers Tagging in Brightspot

Amazon Comprehend is a natural language processing service that uses machine learning to find insights and relationships in text. Brightspot's Amazon Comprehend integration powers our "Suggested Tags" feature, which allows you to match keywords from your article to existing tags in your taxonomy. This saves time for editors, as they no longer need to think of or search through hundreds of possible tags.

As the above demo illustrates, Brightspot’s “Suggested Tags” feature helps make editors’ and reporters’ lives easier.

Once an article is created and you’re ready to add tags, instead of searching through hundreds of possible options, Brightspot allows you to see suggested tags that sync up with the topic of the article. You then simply select the specific ones you’d like to associate with the article.

How is this possible? Well, the “Suggested Tags” feature essentially takes a list of keywords and matches that list up against existing tags within Brightspot. The widget then displays the matching tags as suggestions to editors to add the tags to the article. In this implementation, we’re sending all the text of an article over to Amazon Comprehend, which is able to detect key phrases and entities. We then match the responses we receive against the newsroom’s existing tags in Brightspot. By providing suggested tags in this manner, the CMS saves time for editors, who no longer need to search for or think of relevant tags for each piece of content.

Wondering if a headless or a decoupled CMS is best? Find out what those terms mean, why GraphQL matters and how this CMS technology works.
We're sharing the background story on why Brightspot’s features are the way they are. It’s not just how Brightspot works but why its features were engineered the way they were.

Share
Get the publishing speed you need
with Brightspot.