This project analyzes financial data from the IBM Anti-Money Laundering (AML) dataset to identify high-risk accounts. Using a rule-based approach focused on transaction frequency and total monetary value, this analysis flags potential money laundering suspects by identifying accounts with behavior consistent with structuring.
-Data Preprocessing: Loaded and cleaned the transaction dataset, renaming columns and correcting data types for analysis.
-Rule 1 (Frequency Analysis): Identified accounts with an unusually high frequency of suspicious cash deposits, a common indicator of structuring.
-Rule 2 (Value Analysis): Calculated the total monetary value of these deposits for each account to find high-value targets.
-Data Enrichment: Loaded a second 'Accounts' dataset and merged it with the transaction findings to create a detailed profile of the top suspects.
-The analysis identified two high-risk accounts that met the criteria for both high frequency and high value of suspicious cash deposits.
-These accounts had deposit counts of 24,726 and 14,989 respectively.
-The total value of these deposits was massive, at approximately $22.1 billion and $12.5 billion.
-Both suspect accounts were linked to the same institution: Willows Thrift bank.
-.png)
-Python (Pandas)-for Analysis
-Python (Matplotlib)-for visualization