💳 FraudSimulator: Real-Time Credit Card Fraud Detection Dashboard

An interactive Streamlit dashboard built using Python and the Kaggle Credit Card Fraud dataset. This project simulates real-world fraud detection using both rule-based logic and machine learning (Logistic Regression) with adjustable thresholds and clear visualizations.

---

🚀 Features

• 📂 Upload actual transaction data (creditcard.csv from Kaggle)
• 🧠 Simulate fraud detection using:
  • Rule-based flagging (Class == 1)
  • Logistic Regression model with a probability threshold
• 🎛️ Interactive slider to adjust model prediction threshold
• 📈 Visualize:
  • Log-scaled fraud vs legit bar chart
  • Pie chart showing fraud proportions
  • Confusion matrix for ML performance
• 📥 Download all flagged fraud transactions as CSV

---

📊 Dataset

• Source: Kaggle Credit Card Fraud Dataset
• Rows: 284,807 transactions
• Frauds: 492 (0.17%)
• Features: Time, V1–V28, Amount, Class

---

🛠️ How to Run

1. Clone the repository

git clone https://github.com/YOUR_USERNAME/fraudsimulator.git
cd fraudsimulator