Special Issue on “Agentic AI for Explainable Big Data Exploration and Visual Analytics”

Abstract

The ongoing data revolution is being reshaped by the convergence of Agentic Arti fi cial Intelligence (Agentic AI), explainable data analytics, and visual reasoning frameworks. As Big Data continues to expand across scienti fi c, industrial, and social domains, the ability to autonomously explore massive datasets while maintaining explainability, interpretability, and transparency has become a critical challenge. Traditional data analytics systems, although powerful in computation, often operate as “ black boxes, ” limiting human understanding, trust, and accountability in high-stakes decision-making processes. This Special Issue explores new theoretical foundations, models, and architectures that empower explain-able Big Data analytics through Agentic AI, focusing on interpretable reasoning, knowledge graphs, causal inference, human-in-the-loop learning, and visual data storytelling. It seeks to highlight interdisciplinary research spanning AI, data science, cognitive computing, human – computer interaction, and visualization, with applications across health care, fi nance, smart cities, Industry 5.0, and climate informatics. The fi rst article suggests a Big Data – Driven Person-alized Instruction framework powered by Explainable Agentic AI (X-AI). This work advances a scalable and culturally responsive paradigm for personalized education by combining big data analytics, agentic autonomy, and explainable AI. The authors of the second article have developed an Agentic AI pipeline integrating genome-wide association study (GWAS) data from FinnGen Release 12 (6854 cases; 384,461 controls). GWAS identi fi ed a strongly associated locus on chromosome 19q13.32, with lead variant rs429358 (P = 2.79