In class last week, we examined new ways to visualize literature. This included the introduction to word clouds! Word clouds are essentially visual representations of text, in which words that have used at a higher frequency will appear larger within the visual map. The two main tools we examined were Wordle and Voyant.

I chose to examine “The Speckled Band” through a word cloud to see what type of image the text would produce. I started by imputing the data into Wordle.net, a web based word cloud generator. The generator applies common stop words to the text so common words like “a” or “the” or “and” will not be present in the visualization. I love that this tool allows the user to change the font, direction and color of the visualization! The result:

I found myself underwhelmed with the results. I was largely unsurprised at how large “Holmes” was within the visualization. Given that it was a story about his detective work, this result was expected. Given the story, I was not shocked to see “bed,” “room,” “sister,” and “night” pop up in higher frequencies. One of the disadvantages to Wordle lies in the inability to edit stop words. For example, words like “also” and “ago” popped up in the visualization.

Overall, I found that the visualization did not accurately represent the important aspects of the story. “The Speckled Band” is about a young girl who died mysteriously in her room after hearing a whistling sound and her twin sister hiring Sherlock Holmes to figure out what had happened so that she did not die as well. While the word “whistle” “death” and “died” do appear, they do so at a low frequency and are words that have to be actively searched for within the visualization.

While you could make assumptions about the content of the story, it would be hard to draw active conclusions. I tested this theory by showing the word cloud to several individuals who were not familiar with “The Speckled Band” and asked them to brainstorm what the story may be about. While many were able to pick out that it was a Sherlock Holmes story, which included the death of someone and the dealing with a family, most were completely unable to grasp the important details of the story.

Because of the limitations that are associated with Wordle, I also ran the text through Voyant. This allowed me to remove Sherlock, Holmes, and Watson in order to see how the absence of those key words would effect the visualization. The results:

I found myself even more unsatisfied. The results were almost the same as before, but the words that held little meaning to the story became larger. While key words to the content of the story, like “death,” “dead,” “ventilator,” “bed,” etc. were present, they mean very little without knowing the story. Most of the words present are descriptors of setting, characters or a random assortment of words that have no direct connection to

While word clouds have proven to be visually appealing, I seem to be unable to grasp the applicability of them in the case of “The Speckled Band.” When I picture visual representations of texts, I imagine them relating more deeply to the content of the story. From the word clouds I produce this week, I do not see how this particular type of visualization can relate to the text’s content, though they can be very helpful at analyzing word frequency and the frequency of word type to the right kind of text.