|
Project Information
Members
|
What are Statistically Improbable Phrases? They're phrases that occur more often in a piece of text than in general English. They can give you a sense of what the text is about. The best way to understand this is to try it out. For example, to see which words occur specifically in Calvin & Hobbes, go to http://sip.s-anand.net and type http://s-anand.net/calvin_86.html in the box. You'll see all words in the URL listed alphabetically. The big words occur more often. The dark words are more improbable. This is the source code for the application (hosted on Google's AppEngine) |