History

The Intelligent Learning Experience Lab was started by Dr. Inventado in the Fall of 2018 at Computer Science Department of California State University Fullerton. He collaborated with eight undergraduate students in the development of their first project, the ILXL programming problem repository.

Quick facts:

  • The lab's founding members are Paul Inventado, Leah Smith, Lonnie Hansen, Sunny To, San Tran, Evan Purpura, Andrew Gomez, Ruby Abutaleb, and Dayna Anderson 
  • ILXL's flagship project, the open programming problem repository, was pioneered by Paul Inventado and Leah Smith in Spring 2018
  • The ILXL logo was developed by Ruby Abutaleb in Fall 2018.

Our Research

Research at the Intelligent Learning Experience Lab focuses on four areas to create effective learning and teaching experiences.

ILXL research focus

The first area, Educational Data Mining, involves using machine learning and statistical analyses to understand students’ learning experiences and identify barriers to success. The second area, Evaluation, involves analyzing empirical data and using randomized controlled trials and other statistical techniques to find effective pedagogy. The third area, Educational Design Patterns, encapsulates empirically validated results in educational design patterns. Educational design patterns capture high-quality teaching strategies that address recurring problems in learning that is observed in specific learning contexts. Finally, the fourth area, Learning-software Development, involves developing learning systems that help students learn (e.g., immediate feedback, hints, suggest learning strategies) and support teacher instruction (e.g., student performance monitoring, topic difficulty measurement, semi-automated grading). It utilizes statistical analyses and machine-learned models in predicting student performance and other learning outcomes (e.g., motivation, engagement, frustration) to identify appropriate learning support and suggest teaching strategies that can addresses students’ needs.

Educational Data Mining is used to gain insights on student learning experiences from data generated by learning software and collected from in-class activities (e.g., assessment, attendance, participation). Such insights help uncover effective teaching strategies that are further verified and evaluated using randomized controlled trials. High-quality teaching strategies supported by empirical evidence are encapsulated in educational design patterns that inform the refinement of learning software and teaching pedagogy. This process is repeated so that data generated by refined learning software and teaching pedagogy can be used for incremental improvement.