The Library of Congress is digitizing millions of newspaper pages

Google

Newspapers from old years are a good historical source.

The American Library of Congress came up with the idea of digitizing a catalog of old newspapers and their microfilm. They are a valuable source of knowledge of events of old years, and they are combined to see how people have referred to these events.

Travel through centuries and library of congress and digital newspaper file.

In total, 16 million pages were collected for the entire history of the United States. Digitization uses machine learning techniques.

Scientists have been learning artificial intelligence of identifying individual component texts, such as headings, pronouns, titles, and illustrations, tracing them at the beginning of the frame and then allowing the machine to function.

The system can also organize visual content and distinguish between pictures, illustrations, maps, comics, cartoons and commercials. This allows you to isolate components, making your entire collection easier to navigate. You can search for maps from the time of the selected States, for example.

No censorship

Thanks to the work of scientists, which lasted for 19 days continuously, you can now admire the letters from the years 1789-1963. Of course, with all the good of the inventory: There is also a lot of racist content in the newspapers of that period.

The project will create further collections of this type. Scientists also made newspaper Navigator available for free at this address.