#Project 2

Link to Project: https://github.com/ecbrown7/ST558_project2 Link to Repo: https://github.com/ecbrown7/ST558_project2.git

Project Outline:

THe project was focused on using boosting trees, random forests and comparing them to linear models. It also gave a bit of an exercise in picking variables to model by. The data it focused on was different news categories hosted on mashable.com

• what would you do differently?

I would probably spend more time trying a lot of different models. I’d also like to check how much we gain by each variable and let that help create our model. There were some techniques from ST518 that I really wanted to try out.

• what was the most difficult part for you?

The most difficult part for me was getting an idea of what the data looked like so I could have some meaningful choices on what to touch.

• what are your big take-aways from this project?

My big take away is to establish a template for formatting both with the data and the actual document so that there is no need to go back through and check.

I think my partner and I worked really well together. Evan definitely knew what he was doing with the code and it was nice to use him as a reference for accomplishing certain things.

R to make this post:

#rmarkdown::render(“_Rmd/2022-07-10-Project-2-Blog-Post.Rmd”,output_format = #md_document(“markdown_github”),output_dir = “_posts”, output_options = list(keep_html=FALSE))


<
Previous Post
Coolest Thing I’ve Learned in R
>
Next Post
Favorite Modeling Methods I’ve Learned from ST558