Money-Laundering-Classification

DATASET:
EDA https://www.kaggle.com/code/jeslyjoji/eda-money-laundering-check-on-ibm-transactions
CODE https://www.kaggle.com/code/jeslyjoji/dealing-with-imbalance-data-money-laundering

Models trained on imbalanced datasets tend to focus on predicting the majority class well, often neglecting the minority class. It becomes biased The Dataset is not balanced hence needs to be handled. Used Undersampling technique as the dataset is huge

BEFORE Undersampling

F1-Score 10%

AFTER Undersampling

F1-Score- 89%