While users of the web have documented the past and the present, and put them to the internet, there are several things that may be unique, and are difficult to find.

Those things can include images that explain stories of the past.

To "re-imagine" how people can search for visual content using machine learning techniques, the U.S. Library of Congress, the research library that officially serves the U.S. Congress and is the de facto national library of the country, has released an AI tool dedicated to browse through millions of old and historical newspaper pages.

Called the 'Newspaper Navigator', the tool allows users to see and read seminal events and characters, such as wars and presidents, which have been depicted in the press.

With a wealth of clues about the work of editors that created the look-and-feel of the past news, users can simply go back in time to see history in the making.

This particular tool can be invaluable for journalists or archivists, for example.

The website's about page reads:

Newspaper Navigator was created by Ben Lee, a Washington University researcher and the Library of Congress’ Innovator in Residence.

To make it happen, Lee identified the images using image-recognition AI model, trained on annotations of World War I-era newspaper pages from the Library’s digitized collection of newspapers published between 1900 and 1963.

This allows the AI to detect photographs, illustrations, maps, cartoons, comics, headlines, and even advertisements, printed on old newspaper pages between those years. The AI also uses what its called the 'Optical Character Recognition' to extract the headline of the news, as well as the captions from the articles.

To use the system, users can simply enter a keyword in the Newspaper Navigator, and the AI will surface matches from a dataset of 1.56 million newspaper photos.

The images were taken from Chronicling America, a database of more than 16 million digitized historic newspapers, all of which are in the public domain.

Chronicling America is a product of the National Digital Newspaper Program, and a result of a partnership of the Library of Congress and the National Endowment for the Humanities.