Voyant, Kepler.gl, and Palladio are examples of the many digital tools available to the public for data visualization. These past weeks I have been working with data from the WPA Slave Narratives using different digital tools. Which is the best of those visualizing tools? It depends on what you are hoping to find. Which is the best of those visualizing tools? It depends on what you are hoping to find.
Voyant tool approaches the data from a text-mining and topic-modeling perspective. After looking for repetitive words and considering factors like OCR errors or misspelling, Voyant produces a series of graphs that display the topics more mentioned, in which context, and many other details that would have taken years to accomplish by close reading. And perhaps my favorite feature of them all, the word cloud conveying so much information in such a whimsical way.
On the other hand, I found mapping software very effective in translating text files into historical, memory, cultural, community, and many other maps. Using Kepler.gl was enlightening because I got to experience how efficient mapping tools can be for visualizing data from text files. As words were vital for Voyant, locations are essential for Kepler.gl and all other mapping tools available to the public.
Finally, when the focus of a digital humanities project lies in the relationship between data sets, networks should be the appropriate tool to use. My experience with Palladio was surprisingly smooth; I was expecting that visualizing through networks would be more complex. But Palladio is a user-friendly tool that allows you to rapidly visualize interdependent tables to explore different connections and perhaps ignite new research areas after analyzing the graphs.
The good news is that these visualization tools that I have been exploring in my lasts posts, are not excluding and can be used in combination to each other according to your needs. Going back to the WPA Slave Narratives datasets I used for the practicum activities, I will say that each of these digital tools can be extremely helpful. Each of them has the potential to enhance the existent scholarship by provoking different sets of questions focused on various aspects of the data: narrative’s content, geographical locations, or the relationship among the data.
If I have to choose one, though, I will use Voyant without a doubt. I love words, and text-mining and topic-modeling will be my first choice if I get to be involved in a digital humanities project in the future.