Fundamental Review of Portfolio Project: US Medical Insurance Costs

Projects Portfolio Project: U.S. Medical Insurance Costs
1

Hi Everyone!

I just finished this project and it took me a warping 4:30 hrs! I cant tell if I’ve had breakfast. I was willing to do more but the time is not permitting.

if someone could please check my work and let me know if I’m on the right track. I would tremendously appreciate the 5 minutes.
I enrolled into the Data Science career paths in the hopes to do all of them, starting with Analytics (seemed like the logical choice at the time).
Please understand, my coding style does come across as “compressed” and that is why I use a lot of comments.

Please find my work on Kaggle at Eugene’s US Medical Insurance Costs

2

You don’t need to enroll in all the DS courses. Much of the content overlaps, so when you complete it in one area, it’s marked as complete elsewhere. Also, if you haven’t done so, maybe check out the Python3 course.

Some suggestions & observations:

  • yep, you’re on the right track! :slight_smile:

  • it might be better to just print the first 5-10 rows of data rather than all of it.

  • Initially, some basic descriptive stats would be helpful: how many rows in the dataset, number of men vs. women, smokers vs. non-smokers, people per region, etc.

  • highest and lowest charges is good b/c it shows the range. But knowing that, rather than looking at the mean, perhaps focus on the median of charges b/c there are outliers that pull the mean in the data.

  • Good use of comments. Additionally, you might want to add a brief conclusion, wrapping up your findings.

  • I like the breakdown of the age groups and number of children and charges.

1 Like
3

Greatest review ever written!

My appreciation and some thought processes:

  • I didn’t want to do anything outside of what I’ve learned on the course itself
  • Printing samples of 5-10 rows should’ve been a no-brainer but I’ve honestly never thought of that
  • Basic descriptive stats - I’ll keep that in mind (very useful)
  • Means and medians haven’t been covered although I was tempted to show those

This was an interesting project, and your review made it that even more!

1 Like
4

By “mean,” I meant when you calculated the averages (rather than median)

1 Like