The repository is organized as follows: the jupyter notebook insurance_fraud.ipynb
contains all the code for the project. Each section is marked with relevant headers.
Another notebook in the repository is for a statebins graph in R, it's deceptive that the file ends in ipynb
when the code only runs in R which can easily be remedied
by making an R based google collab notebook.
The full write up for the respository is avaliable on my blog, jabedmiah.
Insurance Fraud Data Analysis and Classification.docx
is an old write-up and is saved in this repository for archival purposes. The entire analysis is performed using Python; the tools I use in the analysis vary from modules such as matplotlib, seaborn, scikit-learn, imbalanced-learn, mlextend, and numpy.