Journal of AI, Data Science and Cyber Systems

Kunal Rathore is a researcher in environmental data science and applied artificial intelligence, with a focus on modeling complex socio‑ecological systems and developing early‑warning indicators for critical transitions. He is currently pursuing his Ph.D. in Environmental Sciences and Applied AI at Oregon State University, where his work integrates dynamical systems theory, machine learning, and ecological forecasting to improve the predictability and interpretability of environmental tipping points. His research is conducted in the Socio‑Environmental Analysis (SEA) Lab, with an emphasis on hybrid modeling frameworks that combine mechanistic understanding with data‑driven inference.

Kunal’s scholarly contributions span explainable AI, feature‑selection methods, and computational ecology, including peer‑reviewed publications and recent work presented at IJCAI 2024. His research interests include early‑warning signal detection, high‑dimensional time‑series modeling, and the development of interpretable machine‑learning tools for environmental decision‑support systems. He has also contributed to interdisciplinary collaborations involving harmful algal bloom prediction, resilience assessment, and AI‑assisted environmental monitoring.

In addition to his academic research, Kunal brings applied experience from industry roles in machine learning and AI systems engineering. His work has included developing forecasting pipelines, anomaly‑detection models, and natural‑language‑to‑database query systems, with a strong emphasis on model reliability, transparency, and real‑world deployment.

Kunal’s broader goal is to advance rigorous, interpretable, and scientifically grounded AI methodologies that support ecological resilience, climate‑adaptation strategies, and sustainable resource management. He actively engages with the research community through conference presentations, invited talks, and interdisciplinary collaborations across environmental science and artificial intelligence.

• 1. Interpretable and explainable machine learning for high stakes scientific and engineering domains 2. Hybrid dynamical–ML modeling for complex system prediction and uncertainty quantification 3. High dimensional time series analysis and forecasting for real world, noisy, and sparse datasets 4. Feature selection and model interpretability for robust, transparent decision support systems 5. Data driven resilience and risk modeling in complex socio technical and environmental systems 6. Applied AI for monitoring, anomaly detection, and early warning systems across scientific applications