Discussion Page - Idea 88

#88: Metrics on Text

How much information can you get from a text, without reading it, just classifying its words?

The idea is to compare a text, any text, with a table where words were sorted by a quality.

Quality can mean many things, They can mean for example if the words are associated with a sentiment (heaven brings good feelings, prison bad ones). It can screen if there are innapropriated words (curses, foul language). Can reflect the erudiction of the author (are there fancy words). Quality can indicate archaism (anyone reading a 18th century novel knows what this is about). Finally, the presence of technical terms from scientific vernacular can indicate technicallity. 

Aynway, the purpose off all this is to have that info displayed on the back of a  book, so you know what to expect before you actually get to read it. Perhaps you are not in the mood that night for old stuff, or you frown on foul words. 

Well, just a thought.

BANNER IMAGE CREDITS: NASA, ESA, Harald Ebeling (University of Hawaii at Manoa) & Jean-Paul Kneib (LAM)