MbShea@PWP@English@UMCP

Beginning:
I will begin with a cover letter explaining the intended audience of the paper, and some of the general topics that the audience would already know. This will be a document separate from the Jupyter Notebook. This cover letter will be written to inform anyone without the knowledge of the target audience, so they can begin reading the document and understand what is being done in very general terms. The cover letter will be ideal for any potential management-type figures with less strictly computer science and statistical knowledge.

Middle:
The middle of my document will include an introduction to the problem statement for this application of predictive analytics and will then move into the actual processes used. This will include scraping for data, fitting regressions, fitting machine learning models and performing K-fold tests. After this, each method, which has been trained on last year’s data, will be applied to the data from this year, to see how we would have performed had we used any of the processes we generated.

End:
The ending of the document will explain the results and the nature of our tests, and why they were used. We will warn of potential problems in the process and how to avoid them. This will be especially useful as people apply the same methodology to their own data sets.

Logos:
We will use sources on the various methods we use to explain them and why they work.

Ethos:
I will provide my background as a Computer Scientists and provide backgrounds for my sources.

Pathos:
While our application does not necessarily display Pathos, the processes used here can be used for a variety of data sets, including ones related to health and medicine. The processes used are very compelling because they can be used to try and generate predictions in whatever data you desire.

Structure:
This will be a Jupyter Notebook, with a cover letter for those who may not have the basic coding and statistical experience recommended to understand the document.

Goal:
To inform about the process of predictive analytics through an extended example, with explanation.

Looking forward to this. In the cover note or the follow-up note, can you describe what problems this type of analysis can be used for?

People reading such strategies often think these applications are linked to betting contexts. Betting is a kind of broader decision making strategy in the face of uncertain information. THIS IS LIFE!

Do you have some resources to refer the reader to? Can they be hot linked in the JN annotations?