I started analyzing the data using DataFrame and visualizations (based on the data types of each column).
Afterwards I noticed that one request was to write code using OOP concepts.
So I extracted some very generic behaviors into 2 functions, where I used DataFrame built-in methods:
- apply transformation on a column
- group by a column and apply transformation on a second column
Visualizations are not part of the structured code, as they were written beforehand and I left them as they were. But I got the most information from data visualization.
I am curios about any opinion on the analytical approach and structure. Probably the overall code structure won’t get a passing grade.
Thank you all in advance!
Here is the code in the GitLab repo: [Project] Medical Insurance.ipynb · develop · Data Science Projects / Medical Insurance · GitLab