Materials for Oct 26th Meetup
Welcome! The goal for today is to introduce data analysis through an applied project. Given the variety in experience in this group, three options are available for you:
Beginner: We will work through this file together, line by line, as I explain reasoning and answer questions as we go along. Experienced: You independently work through appointments_skeleton.ipynb, a file similar to this one but without the answers. Independent: Using the MIMIC2 ICU data, predict mortality or length of stay (https://www.kaggle.com/drscarlat/mimic2-original-icu/version/1). We will reconvene as a group at 7:30 and share what each one did, debrief on learnings, and answer any remaining questions.
Choose your adventure!
Missed Appointments A clinic is in trouble, as they have been wasting resources every month when staff is available but rooms are empty. We have been tasked with understanding why people who receive treatment instructions do not show up at the next appointment time. How often are patients missing appointments? Who is missing most often? What are the contributing factors for missing appointments?
This example is based on a Kaggle dataset and strongly inspired by Veronika Rovnik's blog post.
Blog: https://towardsdatascience.com/exploratory-analysis-python-kaggle-data-b0afb6ec1788