11.14.2015

Wordclouds for lyrics

Word cloud (or a tag cloud) is a very nice way of looking at text data. I'm almost certain that you have seen a lot of them and understand it's principle. But, in any case, I'll put some explanations: most popular words (tags) in the text are taken and placed more or less randomly on a picture such that the more times a word appears in the text the bigger size it has. 

What I did here is that I created R scripts to automatically create word clouds for lyrics archive of any specified artist. Basically, with the help of them one can create a cloud of the most popular words used by this artist. For the source of lyrics I chose metrolyrics.com site. Big thanks to them for having a server that doesn't ban my IP for sending too big amounts of queries. Because at first I tried another site, and they banned me for a day each time I reached five hundreds of queries or so.

Scripts can be found on my github repo Lyrics_WC. If you are familiar with R, everything must be easily understandable, by I put some explanations to the Readme.

Enough of introduction. Here are word clouds for some classic rock bands.
  • The Beatles
  • Queen
  • Rolling Stones
    You can find some other examples at the github repo.



No comments:

Post a Comment