- Article
Integrating Exploratory Data Analysis and Explainable AI into Astronomy Education: A Fuzzy Approach to Data-Literate Learning
- Gabriel Marín Díaz
Astronomy provides an exceptional context for developing data literacy, critical thinking, and computational skills in education. This paper presents a project-based learning (PBL) framework that integrates exploratory data analysis (EDA), fuzzy logic, and explainable artificial intelligence (XAI) to teach students how to extract and interpret scientific knowledge from real astronomical data. Using open-access resources such as NASA’s JPL Horizons and ESA’s Gaia DR3, together with Python libraries like Astroquery and Plotly, learners retrieve, process, and visualize dynamic datasets of comets, asteroids, and stars. The methodology follows the full data science pipeline, from acquisition and preprocessing to modeling and interpretation, culminating with the application of the FAS-XAI framework (Fuzzy-Adaptive System for Explainable AI) for pattern discovery and interpretability. Through this approach, students can reproduce astronomical analyses and understand how data-driven methods reveal underlying physical relationships, such as orbital structures and stellar classifications. The results demonstrate that combining EDA with fuzzy clustering and explainable models promotes deeper conceptual understanding and analytical reasoning. From an educational perspective, this experience highlights how inquiry-based and computationally rich activities can bridge the gap between theoretical astronomy and data science, empowering students to see the Universe as a laboratory for exploration, reasoning, and discovery. This framework thus provides an effective model for incorporating artificial intelligence, open data, and reproducible research practices into STEM education.
15 December 2025





