AI technologies in educational big data analytics demonstrate high predictive accuracy and effectiveness in personalizing learning, but face significant technical, institutional, and ethical barriers that limit widespread implementation.

Objective: The main goal of this systematic review was to comprehensively analyze recent research (2020-2025) on the deployment of artificial intelligence in educational big data analytics, including applications, challenges, and future research possibilities. The study aimed to evaluate the current state of AI use in education, identify key application domains and methodologies, examine technical constraints and ethical issues, and map research gaps to suggest future directions.

Methods: The study employed a systematic literature review methodology following PRISMA guidelines. A comprehensive search was conducted across six major databases (IEEE Xplore, ScienceDirect, SpringerLink, Scopus, Web of Science, and ERIC) using specific search terms related to AI, big data, and education. Initial searches yielded over 1,200 articles, which were reduced to 50 final peer-reviewed journal articles through a three-stage screening process (title, abstract, and full-text review). Quality assessment was conducted using CASP and COREQ checklists, with only studies scoring above 70% included. Thematic analysis was performed to categorize findings into four main themes, with systematic data extraction for each included study.

Key Findings:

• Four primary application themes emerged: predictive analytics for academic performance forecasting (39.7% of studies), personalized and adaptive learning systems (27.6%), learning analytics for teacher support (19%), and institutional decision-making using educational data (13.8%).
• Machine learning techniques, particularly Random Forest and SVM algorithms, achieved high accuracy (88.5% average) in predicting at-risk students and academic performance.
• Adaptive learning systems using reinforcement learning and NLP showed significant improvements in student engagement and retention, with platforms like Duolingo demonstrating 22% higher user retention compared to static content delivery.
• Learning analytics dashboards increased teacher awareness of student progress by 40%, enabling data-informed instructional decisions.
• Technical challenges were the most frequently reported barriers (46.6% of studies), followed by ethical concerns (37.9%) and infrastructure limitations (32.8%).

Implications: The findings demonstrate that AI has transformative potential for education through intelligent data analysis, predictive modeling, and adaptive learning mechanisms. The research provides a comprehensive framework linking AI tools with practical educational applications, showing how technology can enhance rather than replace human teaching. The study contributes to both academic understanding and practical implementation by identifying effective AI techniques for specific educational contexts and highlighting the importance of addressing technical, institutional, and ethical barriers for successful adoption.

Limitations: The study was limited to English-language publications, potentially excluding relevant non-Western research. Grey literature and unpublished works were excluded, which may have restricted the scope of innovative efforts. The rapid pace of technological development means some findings could become outdated quickly. Publication bias may have affected results since only peer-reviewed articles were included. The subjective nature of thematic coding, despite efforts at objectivity, may have introduced some interpretation bias.

Future Directions: The research identifies several critical areas for future investigation: developing scalable AI systems for broader implementation, creating multilingual and culturally adapted AI models to address bias issues, enhancing explainable AI (XAI) frameworks to improve transparency and trust, implementing federated learning and differential privacy approaches for better data protection, exploring integration of generative AI and VR/AR technologies, conducting longitudinal real-world studies to bridge the theory-practice gap, and developing comprehensive teacher training programs for AI integration.

Title and Authors: "Artificial Intelligence in the Era of Educational Big Data: A Systematic Review" by Ali Hussein Khalaf AL-Sammarraie from the Directorate General of Education Diyala, Ministry of Education, Iraq.

Published On: June 30, 2025

Published By: CyberSystem Journal, vol. 2, no. 1, pp. 33-52