EduSuccessX: A Public Synthetic Dataset for Explainable Student Success Prediction in Higher Education

EduSuccessX is a large-scale public synthetic dataset developed to support research in student success prediction, learning analytics, educational data mining, explainable artificial intelligence (XAI), and higher education decision support systems. The dataset has been synthetically generated using statistically grounded data generation techniques to emulate realistic student demographics, academic performance, engagement patterns, behavioral indicators, attendance records, socioeconomic characteristics, and institutional interactions while ensuring complete privacy preservation and the absence of personally identifiable information. The dataset is designed to facilitate the development, evaluation, and benchmarking of machine learning, deep learning, and explainable AI models for predicting student outcomes such as academic achievement, retention, progression, graduation likelihood, dropout risk, and overall student success. By providing a large-scale synthetic alternative to restricted educational datasets, EduSuccessX enables researchers, educators, policymakers, and data scientists to conduct reproducible and privacy-preserving experiments without ethical or regulatory constraints associated with real student data. The repository includes the dataset, data dictionary, metadata documentation, data generation methodology, sample preprocessing workflows, and supporting code for machine learning experimentation. The dataset follows FAIR (Findable, Accessible, Interoperable, and Reusable) data principles and is intended for academic research, educational innovation, benchmarking studies, and reproducible AI research in higher education. Key Features: Large-scale synthetic student records suitable for big data analytics. Privacy-preserving design with no real student information. Comprehensive academic, demographic, behavioral, and engagement attributes. Support for classification, regression, clustering, and explainable AI tasks. Compatible with popular machine learning frameworks and analytical tools. Suitable for research in learning analytics, educational AI, student retention, academic performance prediction, and institutional decision-making. Potential Applications: Student Success Prediction Academic Performance Forecasting Student Retention and Dropout Risk Analysis Explainable Artificial Intelligence (XAI) Learning Analytics and Educational Data Mining Predictive Modeling in Higher Education Benchmarking Machine Learning Algorithms Educational Policy and Decision Support Research This dataset has been released as an open-access research resource to promote transparency, reproducibility, and innovation in data-driven higher education research.