During my experience in the bank's internal audit, a special type of employees who committed more than 3 "super" transactions in their account represent a suspicious behaviour, and could be a very good target to focus more on them to detect the internal fraud.
For this, i created a script which takes as input a file, containing a dataset with informations about the employees, their transactions amount and date.
I use this file Generated_file.xlsx
to feed it to a script, which detects the anomalies (the employees who commited more than 3 transactions), and returns an excel file Montant10Jours.xlsx
, which contains details as the total amount of the transactions, and the dates when the employee commited more than 4 transactions.
For confidentiality purposes, the data used in this notebook isn't real, but was generated randomly using the script generate_random_data.py
.
Text me if you need more details !echo # Automating-Anomalies-Inspection-in-Audit-Detecting-excessive-transactions-