Credit Card Fraud Detection - Data Visualization

This project explores credit card fraud detection through data visualization using Matplotlib and Seaborn. The goal is to uncover insights into fraudulent transactions and their characteristics.

📊 Key Visualizations

1️⃣ Class Distribution: Fraud vs. Non-Fraud

Fraudulent transactions make up a small fraction of the dataset, but their impact is significant.

📌 Key Insights:

• Fraud cases are rare but can involve high transaction amounts.
• Non-fraud transactions dominate, making fraud detection challenging.

📷

2️⃣ Transaction Amounts by Class

A box plot visualizes the spending behavior differences between fraudulent and non-fraudulent transactions.

📌 Key Insights:

• Fraudulent transactions show higher variability in amounts.
• Non-fraud transactions tend to be lower in value and more consistent.

3️⃣ Fraud Activity Over Time

Understanding when fraud occurs can help in building better fraud detection models.

📌 Key Insights:

• Fraud transactions often appear in specific time windows.
• Non-fraud transactions are more evenly distributed throughout the day.