Updated: 8:29 PM UTC, Jul 16, 2017

Recaptcha - An Anti-spam System That Digitizes Old Books

You've probably already used the human verification system reCAPTCHA which helps fight against spam, retyping two words written in a contact form. But did you know that this system allows a clever way to digitize thousands of paper books through your own contribution? Indeed, the two words you type, only one is known and verified.

The other is not known, but by offering hundreds of different users and still having the same response from them, then the system knows that the association between what was scanned and what was typed by users is correct. And that's how we get to digitize thousands of books with an antispam system! The OCR (Optical character recognition) however is not developed enough to be sufficiently reliable. Using the human brain, we get much better results. The slogan of the company (which was acquired by Google) embodies the operation of this system: "Stop spam, read books."


Luis von Ahn: Massive-scale online collaboration
  • Source 1: Web searches

You Might Also Like

Write a comment...
awesome comments!