For the purposes of this assignment, the book I have chosen is The Great Gatsby by F. Scott Fitzgerald because I wanted to analyze how computational text analysis might uncover patterns within a book that revolves around the topics of money, social classes, and the American dream. With the use of a program like Voyant, one can analyze the frequency of the use of particular words, track their recurrence and find patterns within the text.
What makes it particularly interesting about the analysis of literature is that it allows one to study the entire literary piece at once. For instance, it could be possible to analyze the presence of certain words to find out the reoccurring topics like those of money, relations between characters, social position and others. The presence of such words could also show how important these ideas are in specific chapters of the novel. Nonetheless, such trends cannot say anything about the intentions of the author. The presence of words can have something to do with the plot and not the theme itself.
Another approach to analyzing literature could be with the help of AI tools. In contrast to statistics of words used, an AI tool is able to create summaries of each chapter, identify connections between characters, and interpret recurrent themes. However, such use of technology presents new problems because AI tools may generate information and interpretations that could sound true but lack actual evidence. This may cause researchers to miss some important details when working exclusively with AI tools.
The topic of bias and transparency is also important in this age of artificial intelligence. The computer tools are built on certain approaches and principles, therefore, their output will not be neutral. The researchers need to know what the tool can measure and what it cannot measure.
In conclusion, I believe that text analysis through computation is most beneficial when used alongside, rather than in place of, conventional reading. Technologies such as Voyant may help uncover hidden patterns, while AI may suggest topics worth researching. Still, people still have to evaluate the findings, assess the context, and judge if the findings make sense. The point is not to leave all the work to technology, but to enhance our understanding of the texts being analyzed.