There are many different ways to enhance a digital humanities research project and find new information within the project, and the use of text mining, topic modeling and visualization can give further insight into a project. Text mining plays an important role in digital humanities research because it allows people to see the correlations between words that might have a connection to the same topic and, like google Ngram, it can give specific years showing when the word was used often. This type of gathering data gives insight into the connection between relevancy of words over time. While text mining might not be something I entirely use for my own project, topic modeling, another form of gathering data might be a more useful component to my project.
Topic modeling, as Cameron Blevins explained in the the article about Martha Ballard’s diary, is a way of finding words that consistently appear together within text and group them together in hopes that you might find relevancy in words that might not be normally emphasized as important in our eyes. This could definitely prove beneficial to my project since a lot of the language used in the primary documents is not found as often in “modern-day”. Not only will this help my understanding of certain words, but this could also be useful during the time when participating in the slave trade became illegal, yet many still contributed and sailed ships holding slaves. It is very possible that in order to disguise the actual reasons for sailing or the “type” of cargo onboard the ship, captains or the crew might have used different words in writings to send the message of what they were doing without being caught or arrested. Topic modeling may be able to find this possibility in primary documents to add a whole new dimension to my project. Then with word clouds, there can be a visual representation of the most essential parts of the Middle Passage and the slave trade.