Text-Mining are different tools historians and researchers use to gather meaning from a set of information or data. Lincoln A Mullen from America’s Public Bible project states, “The project uses the techniques of machine learning and text analysis to find the quotations, and it uses data analysis and visualization to make sense of them.” (2) Text mining allows analysis of a set of data to gather meaning from it. The types of information a researcher can use can vary. In the case of America’s Public Bible and the Signs@40 projects, researchers use written sources in the form of newspapers and articles, respectively, to gather their information. However, it does not always have to be just written sources. In the case of the Robots Reading Vogue project at Yale University, researchers used visual sources in the form of magazine covers along with written sources to accumulate data. From the data collected, researchers then ask questions and gather meaning from that data. Researchers usually try to find trends and patterns in the data and then gather meaning from those. Historians might want to understand the period or events associated with the information presented, so they would look for any trends or patterns in their collected data to find answers.
Voyant is an excellent text mining tool for historians and researchers to help researchers find those trends and patterns. Researchers can put information into the free online tool and investigate the data using the five tools provided. The five tools work together as a corpus, making it easy for researchers to find the trends and patterns in the information they provide.
When using the tool, I used data from the WPA Slave Narratives collection. When looking at the data on Voyant, I wanted to compare different states and look at the similarities and differences between them. I decided to look at Georgia and Virginia and their most used words with their respective word clouds. Looking at Georgia’s word cloud, I noticed that there were a variety of stereotypical southern words, including plantation, marster, and slaves. The most used word, however, was old.
In comparison, Virginia’s word cloud contained more dialect words, including ah, yo, tuh, and yer. However, the most used word was slaves.
When looking at the articles concerning the word clouds, Virginia focused more on dialect with enslaved people, while Georgia’s was more about life in the South.