Word clouds are a useful way of condensing large amounts of text. Obviously they don’t perfectly describe complex ideas but they can highlight what a text is about in an easy to digest manner. Word clouds typically let the number of times a word is used dictate its size in the cloud. Interesting colours and exotic placement of the words allows for the reader to pick out important terms.
As such I thought I would start the year with a word cloud based upon this blog. What words has this blog focused on?
Doing this was pretty simple:
First I downloaded all the text from the blog into a plain text file.
I next uploaded the text into Mathematica. For those who don’t have Mathematica there are plenty of free to use programs available. Wordle is popular and can be found at: http://www.wordle.net/.
Deleting common words that aren’t especially related to the specific topics discussed allows a clearer focus on the key words in the blog. These include words such as “the”, “at”, “on” and other words that appear commonly in nearly all text not only this blog. These are called Stopwords (which Wikipedia helpfully explains are not to be confused with Safewords). The Mathematica command to use the standard dictionary of words is DeleteStopwords.
Creating the Word Cloud gave the following result.
It turns out that marketing, perhaps unsurprisingly, featured heavily. I’m pleased to say that people also played an important role given their centrality to marketing. Decision, decisions and think are prominent as are consumers, marketers, market and business. I am pretty happy about the result; the coming year will see more of the same.
Read: http://reference.wolfram.com/language/ref/WordCloud.html for details of Wordclouds in Mathematica