Textual analysis is an up and coming tool in Digital Humanities. Topic modeling and textual analysis on the surface seemed to me like math was trying to invade the humanities with graphs and analysis, but through the readings these research methods proved to have some good uses. Topic modeling and text analysis allow for overarching trends in texts or a body of work to become clear. However, to understand these trends, the researcher must have acute knowledge of the text. I think these research methods would be good for narrowing down research topics for a particularly large body of work. I also think that textual analysis would be more favored for literary analysis as the results would seem more easily applicable there, but there are many ways historians could use text analysis and topic modeling. Historians could use this to study patterns in newspapers, letters, government documents, trade records, and much more.
I found the Mining the Dispatch site to be particularly interesting. I liked how the website broke the search down into topics and then subtopics. It was really cool to see how the topic trends changed throughout the Civil War and then compare the topic trends to the events that were happening in the war at that time. I thought that the researchers used topic modeling as a nice compliment to explain the atmosphere of Richmond during the Civil War era. The atmosphere was a changing one and the topic modeling showed that in a more visual way.
Voyant was the tool that I was able to work with to experience more hands on textual analysis. I had some difficulty getting started because an error message popped up every time I tried to proceed into the tool after entering my text. I chose to play around with Voyant using Jane Austen novels from Project Gutenberg. Once I figured out how to remove the common words such as “the” and “and”, I was able to pick out some interesting trends in the way the characters names show up in the story. I liked the feature that allowed you to graph the frequency of a word against the frequency of another word. I used these tools to look at trends in my favorite Jane Austen novel, Pride and Prejudice. After “Mr.” “Elizabeth” was the most common significant word. However, I think my favorite feature was the word cloud generator. It brought in another visual aspect that I liked and thought could be useful in literary analysis especially in the classroom.
Textual analysis and topic modeling is easily applicable to the Bodmer and Maximilian journey project. The journal is absolutely massive so these research methods could help quickly identify not only main themes, but also trends and where to look for further information about those themes. The trends found in the journal will be correlated to where they were on their journey, what tribes they were dealing with, and the time of year. However useful textual analysis and topic modeling prove to be in the study of the Maximilian expedition, I am still a purist and think that researchers should read the parts of the journal that are pertinent to their studies in full detail. Textual analysis and topic modeling are great for an overview of a text or body of work and are a good starting place for research, but a good scholar must always do, as we said in class, “close-reading”.
from Meta Search Alerts copy http://ift.tt/1O4pd5w
Sourced by "The typist writer". The place where writers, bloggers, and publicists come to expand their knowledge in the field content production and publication.
0 comments:
Post a Comment