💳 Credit Card Fraud Detection System

This project implements a Machine Learning-based system to detect fraudulent credit card transactions. Using the highly imbalanced creditcard.csv dataset from Kaggle, a Logistic Regression model is trained and evaluated to classify transactions as legitimate or fraudulent.

---

📦 Dataset

• Source: Kaggle - Credit Card Fraud Detection
• Size: 284,807 transactions
• Fraud cases: 492 (≈0.17% of total)

Note: The dataset is highly imbalanced and contains anonymized features (V1–V28), Amount, and Time.

---

🧠 Model Used

• Logistic Regression
  A supervised learning algorithm used for binary classification (fraud vs. non-fraud).

---

🛠️ Technologies and Libraries

• Python 3.x
• Pandas – for data manipulation
• NumPy – for numerical operations
• Scikit-learn – model building, evaluation
• Matplotlib / Seaborn – visualization
• Imbalanced-learn (optional) – for resampling techniques like SMOTE

---

📊 Evaluation Metrics

Due to class imbalance, accuracy isn't sufficient alone. Other metrics used:

• Confusion Matrix
• Precision / Recall
• F1-Score
• ROC-AUC Curve

---

📂 Project Structure

CreditCardFraudDetection/ │ ├── creditcard.csv # Dataset (download from Kaggle) ├── fraud_detection.ipynb # Jupyter Notebook └── README.md # Project documentation

🔍 Key Steps

1. Data preprocessing and normalization
2. Handling imbalance (e.g., SMOTE / UnderSampling)
3. Splitting into training and test sets
4. Training Logistic Regression model
5. Evaluating with appropriate metrics
6. Visualizing performance using ROC and confusion matrix

---

📉 Sample Output (Confusion Matrix)

	Predicted No Fraud	Predicted Fraud
Actual No Fraud	56,000+	30
Actual Fraud	40	400+

Note: Your results may vary based on train/test split and resampling.

---