search

stat231-s23 / blog1-neuropeeps

Alex Yan & Lily Wickland Shearer
0 stars 0 forks source link

Blog Plan #1

Open alexandyan opened 1 year ago

alexandyan commented 1 year ago

  1. Yes, we will extend our mid-semester project by including unsupervised learning k-means clustering analysis of the freedom vs covid cases scatterplot we generated. In the different clusters, we will pick one specific country and run text analysis on its official COVID policy to see if certain words are related to cases/freedom. For this we will need to include additional data from the official sites of different countries or news sources.

  2. Our blog will work off of our old SHINY App, including our first interactive map displaying individual country freedom and happiness and the scatterplot of covid cases vs freedom, then will add to it depictions of our clustered data (and perhaps tables of them as well). Then we’ll include word clouds from each of our countries COVID policy news sources highlighting key policies.

  3. By 4/20: We want to have our clustered data wrangled. 4/24: Determine countries of interest and find data sources for world cloud data 4/27: Start wrangling world cloud data and work on visualizations for clusters and word cloud, start writeup explaining data. 5/4: Make sure the blog and shiny apps are functioning and all data is displayed, finish writeup 5/9: Finalizing blog and making it look nice

katcorr commented 1 year ago

Sounds good! In terms of the additional data for COVID policy text, I think even official policies changed a lot throughout a given year -- will you try to pull policy text from policies at the beginning of a given year, the middle, the end? Maybe this won't be an issue, but just thought I'd raise it in case you're finding lots of (conflicting?) policies from the same country (I'm thinking, even in the US at a given point in time, policies across states varied a lot). In any case, these are all interesting ideas and will allow you to practice (and show off :) ) your wrangling and analysis skills! If you find you need to cut back and can't produce everything stated here -- particularly since there are only 2 of you -- that's fine, just be sure you include at least the text analysis or the clustering so you're including material from at least one new topic covered in the second half of our semester.

I look forward to your blog post!

Proposal: 10/10