Proceedings of the International Conference on Applied Artificial Intelligence and Emerging Technologies (AAIET'2025)
Description
This book delivers experimentally validated, deployment-oriented artificial intelligence solutions designed to address real-world challenges in energy systems, cybersecurity, health care, optimization, and intelligent computing. It provides benchmarking analyses, comparative evaluations, and scalable architectures that enable researchers and practitioners to move from conceptual models to operational implementation.
The book approaches its subject matter through an application-driven and systems-oriented perspective. Rather than presenting purely theoretical contributions, the chapters emphasize reproducibility, performance evaluation, hybrid modeling strategies, and realistic deployment environments. Each contribution highlights measurable impact, reliability considerations, and integration within distributed and edge-based infrastructures.
What distinguishes this approach is its strong focus on bridging advanced AI methodologies with concrete engineering constraints. The book integrates optimization techniques, intelligent perception models, privacy-preserving learning mechanisms, and cyber-physical system design within unified, implementation-ready frameworks. Special attention is given to explainability, data leakage risks, differential privacy in federated learning, and scalable architectures for industrial and smart infrastructure applications.
The general scope spans smart grids and sustainable energy systems, industrial cybersecurity, metaheuristic optimization, computer vision and remote sensing, healthcare data analytics, graph neural networks for scientific applications, and AI-driven educational technologies. The contributions collectively reflect the multidimensional landscape of modern applied AI and its role in sustainable and secure digital transformation.
The book is intended for researchers, PhD candidates, graduate students, engineers, and practitioners working in artificial intelligence, smart energy systems, telecommunications, cybersecurity, healthcare informatics, and intelligent computing. It serves both as a research reference and as a practical guide for implementing AI-driven systems in complex operational contexts.
Key uses include supporting advanced research, guiding industrial AI deployment, informing system design decisions, and serving as course material for graduate-level programs focused on applied artificial intelligence and emerging technologies.
More details
Content
Digital Twin Driven Unsupervised Anomaly Detection Framework for Cyber Physical Threats in Smart Grids.- Time Series Smart Grid Data Indexing Based on Feature Engineering Using KD Tree.- Improving Satellite Collision Risk Prediction via Physics Informed Generative Adversarial Networks.- Recovery Aware Self Supervised Anomaly Detection for Industrial Control Systems.- SMOTE Data Leakage and Overfitting in Credit Card Fraud Detection Systems A Critical Study.- PanNest A Novel Pansharpening Based on Nested Hierarchical Transformer A Comparative Study of Deep Learning Approaches for Sign Language Recognition.- Lightweight Multi Branch Separable Convolutional Network for Accurate Human Activity Recognition.- Skin Disease Detection Based on Instance Segmentation Using YOLOv12n seg.- Discrete Grasshopper Optimization for Task Mapping in Network on Chip Architectures.- Digital Twin Based Deep Learning Approach for Resource Optimization in MEC Using UAVs.- Hybrid Deep Learning Techniques for Plant Disease Detection.- An Efficient Data Gathering and Analysis in Green Transportation Technology.- Transfer Learning Based Detection and Classification of Photovoltaic Faults Using IR Thermography.