Data-Driven AI for Dynamic Service Ecosystems: From Intent Mining to Proactive Network Assurance

Abstract:

Next-generation digital ecosystems spanning healthcare, manufacturing, IoT, and autonomous systems require real-time, adaptive service delivery. Traditional static infrastructures struggle to meet these diverse demands. This tutorial explores how data mining and machine learning can be embedded into large-scale service networks to enable intent-driven, predictive, and self-healing operations. We introduce the concept of an AI-Native Proactive Marketplace, where enterprise service requirements are captured as structured intents using LLM-based parsing and intent mining. These intents are matched to optimal service configurations through predictive analytics, anomaly detection models, and demand forecasting techniques. A closed-loop framework powered by network digital twins supports proactive assurance, SLA compliance monitoring, and continuous retraining of models.

---

Short Biographies:

Dr. Priyadarsini K is an Associate Professor in the Department of Data Science and Business Systems at SRM Institute of Science and Technology, India. She received her Ph.D. in Computer Science and Engineering and has built a strong academic and research profile in artificial intelligence, machine learning, cloud computing, IoT, and data-driven network systems. Her recently funded project focused on accelerating deep learning models on FPGA for the detection of cardiac arrhythmia, combining medical AI applications with hardware acceleration for real-time performance. She has authored numerous high-impact journal and conference publications, including recent works in the IEEE Internet of Things Journal. She has also co-authored several book chapters on AI for healthcare and IoT data analytics, and has multiple patents published in areas such as cybersecurity, deep learning for agriculture, and telemedicine frameworks. Dr. Priyadarsini actively mentors students and has been recognized with awards such as the Top Mentor Award (NPTEL, 2024) and Excellent Performance Award (IIT Mandi, 2023). Most recently, her team SliceMinds won 1st Place at the ITU FG-AINN Build-a-thon 2025 in New Delhi for their innovative demonstration of an AI-Native proactive network slice marketplace.

Dr. Karthik S is an Associate Professor in the Department of Electronics and Communication Engineering at SRM Institute of Science and Technology, India. With more than 18 years of teaching and research experience, his expertise spans VLSI design, reconfigurable computing, FPGA and ASIC systems, hardware-software co-design, high-performance computing, and machine learning for embedded systems. His research contributions cover both theory and application — from designing advanced transistor architectures such as DG-JL-TFET and FinFETs for analog and RF applications, to bio-inspired optimization in wireless sensor networks and AI-driven IoT system modelling. He has published extensively in leading journals and international conferences, with recent works focusing on energy-efficient transistors, compressive sensing in IoT, and CNN accelerator architectures. Dr. Karthik has also led multiple funded projects, including accelerating deep learning algorithms on FPGA platforms, advanced signal processing for healthcare, and Industry 4.0 automation systems. His work bridges semiconductor device innovation, FPGA-based AI acceleration, and data-driven system optimization, providing a multidisciplinary outlook relevant to next-generation intelligent computing. He has co-authored several book chapters on IoT data analytics, AI–blockchain convergence, and bio-inspired neurocomputing, and has received multiple awards such as the Best Research Award from SRM IST, research colloquium recognitions, and IEEE fellowships for participation in international test conferences.

Dr. M. Prakash is a Professor in the School of Computing, Department of Data Science and Business Systems, SRM Institute of Science and Technology, Tamil Nadu, with over 14 years of teaching and nearly 3 years of industry experience. His research interests include Privacy and Trust Computing, Big Data Analytics, Information Management, and Machine Learning. He has guided several Ph.D. scholars, with three successfully completing their research under his supervision. Dr. M. Prakash has published extensively in SCIE and Scopus-indexed journals, contributing significantly to areas such as cloud computing security, privacy-preserving data publishing, and healthcare analytics. He has been awarded multiple patents, including innovations in privacy-preserving frameworks, AI-assisted systems, and design patents. His funded projects include AI-based agricultural disease detection and deep learning collaborations in image matching. Recognized with awards such as NPTEL Discipline Star and Young Researcher Award, he is also a certified Oracle and Sun Java professional. He has been an Infosys Campus Connect Partner Faculty and WIPRO Certified Trainer, mentoring students for industry readiness.

Trustworthy Machine Learning for Real-Time Operation and Control of Renewable Power Systems: A Knowledge-Guided Approach

Abstract:

On the path towards Net Zero, the growing integration of renewable and inverter-based energy resources is making power systems more variational and complex. Ensuring secure and reliable energy supply now requires faster, smarter operational tools enabled by AI. Because power systems are security-critical physical assets, AI must be trustworthy: its decisions must respect physical laws and security constraints used in operations. Violations can trigger cascading failures, outages, and even blackouts. This workshop investigates how power engineering knowledge and modern machine learning can be fused so that AI aligns with physics-based models for real-time decision-making, with emphasis on safety, robustness, and trust. We will illustrate core operational tasks in power systems, including economic dispatch and stability control, and walk through an end-to-end data-driven pipeline—problem formulation, data curation, knowledge embedding, model design and training, validation, and deployment on benchmark power networks. We expect to develop a shared interdisciplinary toolkit and strengthen links between the AI and power system communities.

---

Short Biographies:

Dr Yuchen Zhang is a Lecturer at the Queensland University of Technology. He received his PhD in Electrical Engineering from the University of New South Wales in 2018 and then joined the ARC Research Hub for Integrated Energy Storage Solutions as a Research Fellow. He was a recipient of ARC DECRA, led projects with power industry partners, and made contributions to multiple IEEE standards in power & energy community. His interdisciplinary team develops and applies advanced AI and optimization technologies to tackle critical challenges in renewable energy transition of power systems. His research interests include data-driven power system analysis and control, trustworthy AI, wind farm planning optimization, and smart cities.

Dr Yuting Zhu is a Lecturer in Electrical and Electronic Engineering at the University of Southern Queensland. She received her PhD from the University of Auckland in 2022 and previously worked at Auckland University of Technology (AUT) and the University of Auckland prior to moving to Australia. She is an active contributor to the IEEE community and engages in interdisciplinary research that integrates artificial intelligence, sensing, robotics, energy harvesting, and optimization for applications in bioengineering and electrical systems. Her work focuses on developing intelligent systems for sensing, soft robotics, data-driven control, and trustworthy AI, with particular emphasis on anomaly detection in power systems and the design of reliable human–robot interaction systems to enhance healthcare, rehabilitation, and assistive technologies.