Trust Model Optimization.pdf

The proposed Weighted Coefficient Firefly Optimization Algorithm (WC-FOA) with Support Vector Machine (SVM) aims to improve trust model calculation and link reliability in cloud computing. The WC-FOA method measures link reliability, while SVM detects malicious users. The model uses entropy measure and link reliability as input to SVM for attack detection. The WC-FOA-SVM model achieves 96% malicious user detection accuracy, outperforming the Random Forest Hierarchical Ant Colony Optimization (RF-HEACO) method with 92% accuracy.

Key features of the WC-FOA-SVM model include:

Weighted coefficient added to FOA to balance exploration and exploitation
SVM used for malicious user detection based on residual energy, bandwidth, entropy, Packet Loss Ratio (PLR), and End to End Delay (EED)
WC-FOA used for secure route discovery
Combination of SVM and WC-FOA to enhance link reliability

The research focuses on addressing challenges and issues related to trust and security in cloud computing, including:

Malicious user detection
Link reliability
Trust model calculation
Secure routing

The study reviews existing trust model techniques, including:

Feedback Fast Entropy-based Detection Strategy (FFED)
Hybrid model combining entropy and Kullback-Leibler (KL)-divergence
Trustworthiness Assessment with Entropy (TAE)

The WC-FOA-SVM model is organized into sections, including:

Introduction to trust management and cloud computing security
Literature review of existing trust model techniques
Explanation of the WC-FOA-SVM model
Results of the WC-FOA-SVM model
Conclusion and future work.