Prof. Dijana Oreški
University of Zagreb, Croatia
Title: Student Success Prediction Based on Machine Learning and Learning Styles
Abstarct: This study aims to develop a robust predictive model for student academic success by integrating machine learning algorithms with learning style analysis. Educational institutions increasingly recognize the value of early performance prediction to implement timely interventions and enhance learning outcomes. The COVID-19 pandemic has accelerated the adoption of online learning management systems, generating vast amounts of educational data suitable for analysis.
The proposed research will extract and analyze comprehensive student interaction data from Moodle learning management system, including course logins, resource access patterns, assignment submissions, and assessment performance. These digital footprints will be combined with learning style assessments to identify patterns of preferred learning modalities and academic achievement.
Multiple machine learning algorithms are applied and compared to determine the most effective predictive model. The methodology includes CRISP DM standard consisting of data understanding, data preparation, model development, and performance evaluation using metrics such as accuracy and reliability.
This research contributes to educational data mining by exploring the intersection between digital behavior patterns, individual learning preferences, and academic outcomes. The resulting model achieves high prediction accuracy, enabling proactive educational interventions that adapt to students' learning styles while leveraging Moodle's AI capabilities for personalized learning experiences.
Bio: Dijana Oreški is an associate professor at the Faculty of Organization and Informatics, University of Zagreb. She obtained her PhD from the Faculty of Organization and Informatics in the field of data science, with a thesis on developing a new method for feature selection. She is the head of the Laboratory for Data Mining and Intelligent Systems (LOUISE). Her research is at the intersection of artificial intelligence and social sciences, focusing on the application of AI and machine learning to solve social problems.
She is the author and co-author of around a hundred scientific papers, some of which have been published in prestigious journals such as Expert Systems with Applications, Applied Soft Computing, Journal of Decision Systems, Computer Applications in Engineering Education, SAGE OPEN, and Artificial Intelligence.
She has worked on several international projects related to artificial intelligence. Currently, she is leading the Croatian Science Foundation project "SIMON: Intelligent System for Automated Selection of Machine Learning Algorithms in Social Sciences", and is part of teams working on projects that apply AI for green energy and environmental problem solving (AI2SEP and OptiSolar AI).
In her teaching, she delivers courses related to artificial intelligence, the application of AI in business, data mining, and intelligent systems.
She is a member of several associations, such as MIPRO, the Croatian Society for Operations Research (HDOI), and serves on the program committees of scientific conferences including ADBIS, MIPRO, and the International Conference on Innovative Computing and Communication.
She has received several awards for her scientific work, including the Faculty of Organization and Informatics Award for Best Scientist and Best Young Scientist, the Rotary Excellence Award, and two best paper awards at conferences.