search

STAT325-S24 / Frankenstein

Other
0 stars 0 forks source link

next steps for simple Shiny app (plus some feedback) #9

Closed jpapagelis24 closed 6 months ago

jpapagelis24 commented 7 months ago

I will use a Shiny app to explore the different narrators in Frankenstein. Here are a couple of the things I could implement. The user would be able to input specific words so that the word frequency can be used in a logistic regression model to determine if a paragraph is in a certain section. When the user chooses a word, they will be able to see the distribution of the word and other EDA graphs, charts, etc., which might be helpful to determine if the word would work well in the model. The user can also have other variables available to use along with EDA, such as sentence length or paragraph length. I could randomly split the novel into a training set and a test set so that the user can see the accuracy of their model and adjust it to their liking.

nicholasjhorton commented 7 months ago

This looks like a good plan. I'd suggest starting with something that lets you visualize some or all of the book.

Please don't hesitate to reach out if I can be of assistance.

nicholasjhorton commented 7 months ago

Might you be willing to share some ideas (perhaps hand-sketched and scanned?) Let's check in briefly about how this might look.

nicholasjhorton commented 6 months ago

Any updates on this front? It may be worthwhile to stub out something simple (that loads your data and displays it in a rudimentary manner) as a way to start brainstorming a possible interface. As always, please let me know if you have questions or run into any issues.

jpapagelis24 commented 6 months ago

@nicholasjhorton Plan for Shiny Analysis: I have decided to switch up the direction of the app and instead of implementing a logistic regression model, I will instead use cleanNLP to examine the types of words (nouns, verbs, etc) that are located in sections of the novel, perhaps using a graph visual. Maybe have mutliple tabs that can used to explore the different parts of speech. I will just focus on verbs and nouns for now.

Preliminary steps to accomplish: 1) Split chapter 24 into two sections (each representing the two narrators) 2) Make the narrator a variable 3) Wrangle the data into a version that Shiny app can use so the user can explore 4) Create a visual

nicholasjhorton commented 6 months ago

This is coming together nicely: I now have a glimpse for what you are hoping to build.

Please see https://github.com/STAT325-S24/Frankenstein/commit/e713589d26002a6149f7a391deca6f38635cdd0c for minor tweaks I made so that I could run the app.

Note that #12 needs to be fixed so that the annotation data is coming from the Frankenstein package.

The app is off to a good start. It will be important to:

  1. allow the user to see some of the text
  2. have an "About" tab that tells something about you and the book

I'd be glad to take a look during class tomorrow as I'm not sure exactly what to make of some of the displays.