Wikipedia-Scanning Bot Suggests Popular Public Domain Reads

By Gary Cutlack on at

Not sure what public domain book to try next now you've given up on Moby Dick for the 23rd time? An algorithm that scans Wikipedia and attempts to rate the popularity of authors might be able to help suggest the next impenetrable copyright-free classic to whack on your Kindle.

The system, built by Dartmouth College's Allen Riddell, scans the Wikipedia entries of over one million authors, ranking them by popularity, length, revision dates and more, with an aim to pick out the most popular choices and create a Public Domain Rank for each writer.

This data is then applied to a list of over a million public domain books already available online, creating a sortable list of reads people are most likely to like based on author popularity. The aim is to allow book digitisers to concentrate on the most popular releases from authors the public care about when the copyright of titles lapses. [MIT via Engadget]

Image credit: Books from Shutterstock