This is a prototype; please don't share it over social media until I can figure out how it scales under load. If it's broken, feel free to e-mail me.

Instructions

Cut and paste any text into the box below. Or use one of the sample texts.

Once a suitable text A list of relevant texts will appear below, selected from the 4.5 million public domain works in Hathi based on how closely their vocabulary matches your query. In my testing, this returns extremely relevant texts--usually copies of the same work first, followed by other texts with a similar subject matter.

The resulting matches are useful in a wide variety of application: building up a personal workset for HathiTrust, for example, or finding books that are similar to any reference text. You could use this same technology to build a browser plugin that would find related books for any news article you were reading on the Internet.

Pre-loaded texts

Here are some demo texts to load. Click on the title (or the blue button to show more examples). But anything will work that has likely close partners in 19th century books, in any language. (Well, any alphabetic language. If you speak Chinese or Japanese, let me know how it works.)

Huckleberry Finn turns up books by Twain and intriguingly similar boy's fiction.


The Hamlin Garland short story "Under the Lion's Paw" turns up the book it appeared in and others by the author.

William James's 11th "Talk to Teachers" turns up a variety of informal works on psychology, education, and attention.

The second section of the Communist manifesto, in German, returns German communist literature (some of which quotes the original).

A messy copy of Tolstoy's Kreutzer Sonata, in Russian, works very poorly (because of orthographic changes?)

William Jennings Bryan's "Cross of Gold" Speech returns it and similar political speeches.


Paste your text here

Similar texts