Malicious Nodes Detection with ANN

This repository contains the implementation of Malicious Nodes Detection based on Artificial Neural Network (ANN) for Internet of Things (IoT) environments.
The work was originally presented at the 22nd International Conference on Computer and Information Technology (ICCIT 2019).

Although this project was developed in 2019, it remains publicly available for reference, learning, and citation.

Overview

The goal of this project is to detect malicious nodes in IoT environments using an Artificial Neural Network (ANN) model.
The ANN was trained to differentiate between normal and malicious nodes based on device behavior features.

Key Features

• ANN-based classification of IoT nodes as normal or malicious.
• Feature selection focused on IoT device behavior.
• Early exploration of machine learning applications in IoT security.

Project Structure

• Malicious_Nodes_Detection.ipynb: Jupyter Notebook containing the complete ANN implementation and evaluation.
• Dataset files: (if applicable, mention here or provide dataset download instructions).
• Supporting scripts and resources.

Requirements

• Python 3.x
• TensorFlow / Keras
• NumPy
• Scikit-learn
• Jupyter Notebook

You can install the required packages using:

pip install -r requirements.txt

Citation

If you use this work or find it helpful, please cite:

@INPROCEEDINGS{9038563,
  author={Khatun, Mirza Akhi and Chowdhury, Niaz and Uddin, Mohammed Nasir},
  booktitle={2019 22nd International Conference on Computer and Information Technology (ICCIT)}, 
  title={Malicious Nodes Detection based on Artificial Neural Network in IoT Environments}, 
  year={2019},
  pages={1-6},
  keywords={Internet of Things (IoT); Machine Learning; Artificial Neural Network; Device Type Identification},
  doi={10.1109/ICCIT48885.2019.9038563}
}

---

Author: Mirza Akhi
Affiliation: University of Limerick, Ireland