Assignment 10A Approach

Approach

For this assignment, we will be using sentiment analysis to experiment with a tokenized dataset. Using chapter 2 in our tidytext book as an example, there is some neat Jane Austen sentiment analysis in there. We will be bringing in this analysis and referencing it within our document.

With this we will expand our analysis by looking at a different set of texts and a different lexicon. I have thoughts on how I would want to approach this, but I am not sure what is legal. I have recently been into the Dune books, so I would like to perhaps use these as a text dataset to perform sentiment analysis. From there, I want to use a sentiment lexicon that is focused on sci-fi. I may look for one that already exists, but i may also see what i can create with Claude, I feel that this could be an interesting application to see if Claude could generate a new rating lexicon with similar parameters to another lexicon. I would like to see how the mood over books changes based on the words used, and potentially use a “strengths circle” chart to visualize the results I am interested in. My backup option will be to look for some other series of public access literature to potentially try this visual on, and maybe continue down the sci-fi Claude lexicon as well.