Cipher Algorithm Detection

This repository contains a machine learning-based approach for identifying cryptographic algorithms from byte sequences. It utilizes various deep learning models, including CNNs, RNNs, and ensemble models, to classify encrypted data.

📌 Features

• Supports multiple cryptographic algorithms (e.g., AES-256, DES, RSA, etc.).
• Implements CNN, RNN, and ensemble models for classification.
• Uses Jupyter notebooks for data processing, model training, and evaluation.
• Pre-trained models included for quick testing.

💁️ Repository Structure

├── README.md                     # Project documentation
├── dataset.ipynb                  # Dataset exploration and preprocessing
├── aboutdataset.ipynb             # Additional dataset insights
├── ensemblemodel.ipynb            # Ensemble model implementation
├── cnnmodel.ipynb                  # CNN-based classification
├── model.ipynb                     # General model training script
├── modelASYMMETRICCNN.h5           # Pre-trained asymmetric CNN model
├── modelasymmetricrnn.h5           # Pre-trained asymmetric RNN model
├── modelcnn.h5                     # Pre-trained CNN model
├── modelstreamcnn.h5               # Pre-trained streaming CNN model
├── modelstreamrnn.h5               # Pre-trained streaming RNN model
├── rnmmodel.h5                     # Pre-trained RNN model
└── rnstream.keras                  # Serialized Keras model

📊 Dataset

The dataset consists of byte sequences paired with their corresponding encryption algorithm labels. It helps the model learn patterns that distinguish different cryptographic methods.

Example data format:

{
  "byte_sequence": "8b597365180634...",
  "algorithm": "AES-256"
}

🛠 Installation

1. Clone the repository:

   git clone https://github.com/Narayanan1411/cipher-algorithm-detection.git
   cd cipher-algorithm-detection

2. Install dependencies:

   pip install -r requirements.txt

🚀 Usage

Run the Jupyter notebooks to explore the dataset and train/test models:

jupyter notebook

Open ensemblemodel.ipynb or cnnmodel.ipynb to start training.

📌 Future Improvements

• Expand the dataset with more encryption algorithms.
• Improve model accuracy with hyperparameter tuning.
• Deploy as a web API for real-time classification.

🐜 License

This project is open-source. Feel free to contribute!