Skip to content

AnisZakari/detect_fraudulent_activities

Repository files navigation

Detect Fraudulent Activities

OBJECTIVES

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

Worflow

  • Exploratory Data Analysis
  • Data Visualization
  • Feature Engineering
  • Data Preprocessing
  • Supervised Machine Learning - Logistic Regression
  • Supervised Machine Learning - Decision Trees and Random Forest