This record contains two datasets and three Python scripts used for analysis in the PeerJ article "SDN-enabled adaptive security framework for multi-cloud infrastructures using deep learning-based threat detection and policy management"

PEERJ-DATA Supporting Code and Datasets for Software-Defined Multicloud Defense Controller (SDMDC) This repository contains the complete implementation of the Software-Defined Multicloud Defense Controller (SDMDC) along with the datasets used for training and evaluation. The framework integrates Local Intrusion Detection Systems (LSTM), Global Intrusion Detection Systems (CCTT Transformer), and policy optimization (Lemurs Optimizer) for enhanced multicloud security. 📂 Folder Structure 1️⃣ Code Contains the source code for SDMDC, organized as follows: Global IDS/ cctt-with-lo.py – Implements the Global IDS using a Cross-Cloud Threat Transformer (CCTT) combined with the Lemurs Optimizer for dynamic policy adjustments. Local IDS/ local-ids-using-lstm.py – Implements a per-cloud Local IDS using LSTM for anomaly detection in ingress and egress network traffic. sdmdc-controller.py – Main execution file that integrates the Local IDS, Global IDS, and policy optimization into the SDMDC framework. 2️⃣ Datasets Contains datasets used for training and evaluation: BOT-IOT/ Description: IoT network traffic dataset designed to detect botnet attacks and anomalous activities. Source: Publicly available via Kaggle: Bot-IoT Dataset (originally uploaded by Vignesh Venkateswaran). CICIDS/ Description: Comprehensive intrusion detection dataset with realistic network traffic and various attack scenarios. Source: Publicly available via Kaggle: CICIDS Dataset (uploaded by Rashmi Verma). 🚀 How to Run Unzip both folders (Code.zip and Datasets.zip) in the same directory. Ensure you have the required Python libraries installed: bash CopyEdit pip install numpy pandas scikit-learn tensorflow torch Run the sdmdc-controller.py file to start the integrated SDMDC framework: bash CopyEdit python sdmdc-controller.py Modify file paths in the scripts if datasets are stored in a different directory. 📄 Notes All code in this record is authored by Rashmi Verma and Manisha Jailia. Datasets are included for research and educational purposes. Ensure adequate computational resources for training large models (GPU recommended).