International Journal of Sustainable Development in computer Science Engineering

Explainable AI in Healthcare: A Review of Interpretability Techniques and Applications

Prof. Bhim singh — 2025-01-02

Artificial Intelligence (AI) and Machine Learning (ML) have significantly transformed healthcare by enabling predictive analytics, early diagnosis, and personalized treatments. However, the black-box nature of many ML models raises concerns about their trustworthiness and interpretability. This review explores various explainability techniques, including SHAP, LIME, and attention mechanisms, and their applications in healthcare domains such as radiology, genomics, and clinical decision support. We discuss challenges in implementing explainable AI (XAI) and propose future research directions to enhance model transparency while maintaining predictive accuracy.

Federated Learning in Edge Computing: Challenges, Security, and Future Directions

Dr. Prakash Singh — 2025-01-03

Federated Learning (FL) has emerged as a promising paradigm for training ML models across distributed edge devices while preserving user privacy. This review paper provides an in-depth analysis of FL architectures, optimization techniques, and security challenges, including adversarial attacks, model poisoning, and data heterogeneity. We explore real-world applications in IoT, healthcare, and smart cities and discuss future advancements to improve scalability, efficiency, and robustness in FL frameworks.

Ethical Considerations in AI-Driven Decision-Making: A Systematic Review

Prof. Singhania Shanrma — 2025-01-07

As AI-powered decision-making becomes increasingly prevalent in domains such as finance, healthcare, and governance, ethical concerns regarding bias, accountability, and transparency have emerged. This paper reviews existing literature on ethical AI frameworks, fairness-aware ML algorithms, and regulatory guidelines to mitigate biases and ensure responsible AI deployment. We analyze case studies of ethical AI failures and discuss strategies for designing equitable and socially beneficial AI systems.

Advances in Reinforcement Learning for Autonomous Systems: A Survey of Techniques and Applications

Sri Raja Krishnamurthy — 2025-01-02

Reinforcement Learning (RL) has revolutionized the development of autonomous systems, including robotics, self-driving cars, and smart grid management. This review provides a comprehensive overview of RL algorithms, from traditional Q-learning to deep reinforcement learning approaches like DDPG, PPO, and SAC. We examine applications of RL in real-world environments, address challenges such as sample efficiency and reward design, and highlight future directions for improving generalization and safety in RL-based systems.