Credit Card Fraud Detection using ANN is a sophisticated deep learning solution aimed at identifying fraudulent credit card transactions. Utilizing Artificial Neural Networks (ANN), the model meticulously analyzes transaction patterns to differentiate between legitimate and fraudulent activities. The system effectively addresses the challenge of imbalanced data through techniques like SMOTE (Synthetic Minority Over-sampling Technique) and offers a user-friendly web interface via Streamlit for real-time predictions.

Key Features

• Deep Learning Model: Built using TensorFlow/Keras with a Sequential ANN architecture.
• Data Preprocessing: Implemented StandardScaler for feature normalization.
• Imbalance Handling: Used SMOTE to balance the dataset.
• Interactive UI: Deployed on Streamlit for easy user interaction.
• Real-time Prediction: Classifies transactions as "Normal" or "Fraud" instantly.

Technologies Used

• Programming Language: Python
• Deep Learning Framework: TensorFlow / Keras
• Libraries: Pandas, NumPy, Scikit-learn, Matplotlib, Seaborn
• Deployment: Streamlit
• Tools: Jupyter Notebook / VS Code