For my text analysis adventure, I chose Wuthering Heights by Emily Brontë and used Voyant Tools to explore the novel through computational text analysis. I was interested in this novel because it first appeared in my mind. Although reading a novel closely helps us understand its characters and themes, computational tools offer another way to examine a text by identifying patterns that might be difficult to notice during ordinary reading.
I started the exploration by opening the online text of the book, then copy and paste the URL into the Voyant tools. Then, the first feature I focused on was the word frequency visualization. The word that appears most frequently in the book was “heathcliff” and “catherine”, which are both names for the characters. This is unsurprising as the story revolves around relationships. However, by only staring at the frequency of the words, it is hard to get a sense of the main theme in the book. Therefore, what I tried to do next is look into the overall distribution of particular names in the books, but it doesn’t review much unless I connect those to the things happening in the book.
This shows that the computational text analysis may be useful somehow, but there are some limitations for it. Voyant Tools can count words and display patterns, yet it cannot automatically explain the meaning behind every pattern. For example, the word “love” may appear in the text about a loving conversation or a formal argument. Without combining with the text itself, the data is meaningless and misleading. This problem can be especially important in the age of AI. While large language models help summarize large amounts of text, identify themes and connection between characters, the result it provided can also be misleading and over simplified. Instead of automatically trusting the tool, we should check the essential part by ourselves. In the next few weeks, I may try to analyze the data obtained from Voyant tools by connecting it with the narrative of the story.