E-commerce websites are more and more used, some e-shops have to handle thousands or even millions of transactions a day. This opens room for potential fraudulent activities like money laundering or, use of stolen credit card etc.
This Notebook aim's at computing the probability of a transaction being fraudulent thanks to Machine Learning.
Datasets :
Fraud_Data.csv
IpAddress_to_Country.csv- Exploratory Data Analysis
- Data Visualization
- Feature Engineering
- Data Preprocessing
- Supervised Machine Learning - Logistic Regression
- Supervised Machine Learning - Decision Trees and Random Forest