Text Mining


I  selected 100 words and counted their frequency for each tour.  Since the Western Tour is a lot longer than the others, it often has the highest count.  Looking at the percent within each tour gives a fairer picture. The search for words is a little imperfect.  It will count "Bear" as the same thing as "Bearberry" and "Butt" and "Butter" alike.  I'll be working on a better version.

Click on the headers to sort:


And here are words clouds made from these numbers in an application called Wordle:

Northern Tour

Eastern Tour
Western Tour
And treemaps which also incorporate the categories.  Hover for numbers (why the are 2 I can't say).  Since the charts are the same size, proportions are correct while numbers are different.

No comments:

Post a Comment