For my textual analysis, I chose to explore a book I recently finished titled Little Women by Louisa May Alcott. Little Women seemed like a good choice, as I remember the text well enough to evaluate some of the analysis forms based on prior knowledge. After downloading the novel’s .txt file, I tried out two of the three methods using Voyant tools on its own and Spyral. The capabilities of each form of exploration were fascinating, especially when they allowed me to combine multiple ideas from the week. Let me begin by discussing what I found using various Voyant tools’ display options.
Cirrus Display of Frequent Words in Little Women
Frequency of the March Sister’s Names in Little Women
We had explored the Cirrus display in previous classes, so I began there and realized, while looking at it, that character names were quite frequent. To explore that element more, I chose the Bubblelines tool to compare the frequencies of the sisters’ names in the book. After studying more tools, I moved on to using Spyral to engage in coding-based exploration of the text. Early on, I created code that did a multi-component analysis:
myCorpus.tool("corpusset",{"panels":"cirrus,reader,trends,summary,contexts"});While learning to use Spyral effectively, I followed the sample notebook and further explored the text. This exploration included creating trendlines of character name frequency, identifying various ways words appear together, and a links display that shows the connection between roles and is interactive through movement. I also explored one of the optional Network Analysis tools, Paladio. To figure out how to use Paladio, I similarly followed a tutorial presented by Marten Düring.
Paladio Network Analysis Example
The Network Analysis tool would be handy, particularly for analyzing relationships between people, which I could see applying to many kinds of projects. People in the field of literature might use it to map the relationships among characters in books, or historians could use it to display the relationships among nations.
After spending time with various tools, something I am worried about in this “Age of AI” is losing the context that applies meaning to the text within these tools. As a human user, it is easy for me to pick out various issues in the display compared to what I intend, which I do not see AI being able to do at the moment. However, AI is quickly adapting, so it could bridge this gap soon.