index : projects : What the Dickens?

Sorry! This page requires JavaScript.

Is your writing more like Charles Dickens or Jane Austen? Enter text into the box and click "Guess" to find out. For best results, try a paragraph or more.


What/How

It uses Bayesian classification, the same technique used by most spam filters. It splits the text into words and counts them, then examines its corpus to find out if that word is more likely to occur in Jane Austen (for reasons of space, just the contents of Pride and Prejudice) or Charles Dickens (roughly half of Great Expectations).

Its only purpose is as a useless toy application with which to test my JavaScript port of Divmod Reverend, a Python tool for naïve Bayesian analysis.

Problems

Only tested in IE6 and Firefox 2. Confirmed not to work with IE Mac or Safari 1.3.