Thursday, 29 March 2012

Ngram tracking words over time

Part of Google's mission in scanning millions of books is to extract the information and make use of the knowledge. The extraction and analytical process is well described in the video from TED "What we learned from reading five million books". There is a Google books service that allows public to make use of the scanning tools to look over the text of the scanned books.

Time to have some fun, and see how some trends move over time in the text of published books.

First we have cat (blue trace) and dog (red trace). Looks like "Dogs rule and cats drool" after all.

















Next we have some moving trends that show the changing balance over time.
Here we have canary (blue) v goldfish (red). Big decline of Canaries in the 1950s for some reason.






Finally rat(blue) and mice(red). From the neon-zero trace all the way back to 1800 you just know these have been a persistent pair of pests. The big kick up in the 1980 possibly coincides with the fast food generation. In the last few years up to 2000 mouse starts to kick ratty ass. That change over may of course be the new more electronic meaning of mouse starting to have an influence.


Try it for yourself see what insight you can find at http://books.google.com/ngrams

No comments: