Early on in my usage of Voyant tools, I ran into issues installing the “Desktop version” of the tool to my MacBook. While the GitHub and the small blog posts were useful of the installation of the app, I did not have Java on my MacBook, and even had to move a few files/folders to the “applications” page to truly use the closed server version of Voyant. After some quick and easy trouble shooting with Dr. Seefeldt, and moving the Voyant folder into my applications folder, I was easily able to access the “desktop” version through the web, which allowed me to hosue my own server and not use anyone else’s or lose any data.
I had to redo this one, I ended up deleting the original screen shots of the tool exercise from class. I think this was, however a happy accident. Without this retouch, I wouldn’t have gotten the full corpus of what the tool can do. I went into it the second time, with articles for research I was conducting for my old south and slavery course, and this new mindset allowed me to use the tools in a more intuitive way. Instead of being virtually overwhelmed by the busy interface (which I still believe is incredibly busy) I chose visualization tools that actually helped me see trends, instead of what I did before which was fully a choice in which of them looked prettiest and complimented the corpus of my reading responses for the semester. These pictures are listed below with some brief descriptions.
the image above is a screenshot of what the tool looks like without any manipulation, I used a corpus of papers discussion white women of the old south and their interactions/dynamics with slavery.
The tool above may be my favorite. Its called the mandala. In this tool you see the overall most sued words on the outside, and the individual sources on the inside. When you highlight one source or topic, the strings connecting them highlights as well, showing the overall connections in topic between your corpus sources.
The two pictures above are a before and after manipulating the data. As I accessed the sources from a Clemson University library system and through a Clemson University research tool, the pdfs often have a watermark. Because of this, I chose to show this visualization where you can deselect the words you know are not truly apart of your corpus, but may be used over and over. I replaced Clemson with the word white, as the distinction between white women in this corpus of literature is incredibly important.
On the right, above, is the dotted line that is separated by word, and to the right, it is separated by source, and in each source the words are jumbled together. I could see this specific tool being useful in that you can see the absolute difference in visualization here when you manipulate it.
In the question of whether I would use Voyant again, the answer is most likely. I don’t know if I would use it for a formal research project much in the future, unless I was truly doing a study focused on language or syntax, but I see it being useful for creating a visualization to help prove a point that is already proved with another type of arguement ro as supplemental.