Attribute-Based Adaptive Homomorphic Encryption for Big Data Security.

Area of Science:

• Computer Science
• Cybersecurity
• Data Science

Background:

• Rapid global internet and mobile phone penetration generate massive datasets, commonly known as big data.
• Ensuring the security and privacy of this big data is a critical challenge in its management.
• Existing cryptographic techniques may not sufficiently address the unique security demands of big data.

Purpose of the Study:

• To introduce Attribute-based Adaptive Homomorphic Encryption (AAHE) for bolstering big data security.
• To integrate Oppositional Based Black Widow Optimization (OBWO) for optimizing AAHE key parameters.
• To enhance the convergence and efficiency of the Black Widow Optimization (BWO) algorithm.

Main Methods:

• Developed an AAHE methodology incorporating OBWO for optimal key parameter selection.
• Implemented encryption and decryption processes within the AAHE framework.
• Evaluated the methodology using non-abelian rings and ciphertext homomorphism, focusing on conjugacy examination for one-way security.

Main Results:

• The proposed AAHE with OBWO demonstrated superior performance compared to conventional cryptography techniques like RSA and ECC.
• Evaluations using adult and Microsoft web datasets showed significant improvements in encryption/decryption times, key size, and processing efficiency.
• OBWO outperformed BWO, Particle Swarm Optimization (PSO), and Firefly Algorithm (FA) in key generation processes.

Conclusions:

• The developed AAHE methodology effectively enhances big data security and privacy.
• OBWO significantly improves the efficiency and convergence of key parameter optimization for AAHE.
• The proposed system shows strong potential for real-time application in securing big data environments.