fraud-detection-handbook
fraud-detection-handbook copied to clipboard
Reproducible Machine Learning for Credit Card Fraud Detection - Practical Handbook
Small note: the link to the Forword is incorrect. `.html` should become `.md` maybe? I think the book could use an Abstract or Management Summary containing the academic or business...
The parallel_apply function in customer_profiles_table[ 'available_terminals' ] uses `get_list_closest_terminals`, a function which doesn't exist. Although its a comment, I figured the correct function would be `get_list_terminals_within_radius`.
Congratulation!. This is the best work i have found for fraud detection and i hope you keep up. I have noticed the function is_night is not working as expected due...
The color code used in the example plot of terminals within a customer's radius does not match with the text (red and blue are inverted).
The subsection `Transaction generation process` appears in the section `2.1. Customer profiles generation` but it seems more logical to me that it should appear in the earlier section `Design choices`.
The line: ``` transactions_df[transactions_df.TX_FRAUD==1].TERMINAL_ID.iloc[0] ``` Fails if the transaction with an index of `0` within `transactions_df` is not identified as fraudulent. I also updated the code from `transactions_df.TX_FRAUD==0` to `transactions_df.TX_FRAUD==1`...
I fixed these previously but forgot to open a PR back then, hopefully this isn't too late!