Network Meddling Detection Using Machine Learning Empowered with Blockchain Technology.

Area of Science:

• Computer Science
• Network Security
• Machine Learning

Background:

• Real-time network data analysis is crucial for identifying malicious activities.
• Traditional meddling detection methods often face challenges with accuracy and efficiency.
• Advancements in machine learning offer new possibilities for robust anomaly detection.

Purpose of the Study:

• To develop and present a non-faulty framework for analyzing and detecting meddling in real-time network data.
• To identify diverse meddling patterns detrimental to communication systems, academic institutions, and industries.
• To leverage machine learning for early-stage detection and prevention of meddling activities.

Main Methods:

• Implementation of a framework utilizing data collection and processing techniques.
• Application of machine learning algorithms, specifically Support Vector Machine (SVM) and K-nearest neighbor (KNN).
• Integration of blockchain technology to ensure data privacy and model protection during meddling detection.

Main Results:

• The proposed framework demonstrated high detection accuracy (DA) and low misclassification rates (MCR).
• Support Vector Machine (SVM) achieved a training DA of 99.59% and MCR of 0.41%.
• SVM achieved a testing DA of 99.05% and MCR of 0.95%, indicating superior performance.

Conclusions:

• The developed framework provides effective real-time meddling detection capabilities.
• The integration of SVM and blockchain technology enhances network security and data integrity.
• The findings offer significant benefits for securing various communication and transaction processes against meddling.