Seeing Patterns

Why Visualize Data? It can be very useful for seeing holes in research, and also for analyzing data further to understand how a dataset is interconnected. This article by S. Graham, I. Milligan and S. Weingart  explains the role of visualization in research. There are many different tools to approach visualizing data, here I will be using Voyant. This tool can read either cxv or txt files. If uploading a folder of ordered text files, Voyant will visualize the data in chronological order. This allows you to see the changes in word frequency and use over time.

image

I like how Voyant visualizes the text in several ways so you can compare the frequency of the words using different methods. I do find, however, that in comparison Overview was simpler to use as I had to take more time to understand all of the Voyant features I needed. As part of the exercise, I tried the RezoViz tool to visualize the relationships between words in the URL that was provided but the system generated an error. So I decided to try a URL of my choice, I used a Government of Canada article “Canada’s Action Plan on Open Government 2014-16.” The results were interesting, I did notice however that there were some terms missing that I would have expected to see visualized, such as “Open Government,” “access to data,” or “open data.”

image

I also tried the Stopword Tool which is used to tell Voyant to ignore words with less meaning such as in, from, to, etc. There is a pre-existing stop word list, you can add to this list, also remember to select the options icon, then select Taporware (English) for the Stopword tool to be applied. I tired this with the URL for /shawngraham/exercise/gh-pages/CND.csv and the results were very clear. Voyant removed the words to, and, in, etc. to display the frequency of words with more relevance:

image