Evolutionary Algorithm Optimization of Interactive Factors in Inquiry-Based Learning Environments

• In Event: 61.069 - MTCC Roundtable Session 24
  In Roundtable Session: 61.069-9 - Data Analytics and Research Methodologies in Cognition and Learning

Abstract

There is strong interest in improving the quality of individualized learning support in multimedia environments through automated assessments and personalized interventions. In this study, we employed datamining techniques to develop learner models of students interacting with a map to make observations and inferences during scientific inquiry. Log trace data was collected from 143 middle school students in an effort to identify and optimize predictive factors of success during a phenomena-based, online scientific investigation. Evaluation of the optimized parameters and subsequent decision tree show that these techniques were effective in developing a model that can predict how student interactions with an interactive map reflect different levels of understanding and reasoning with scientific observations.

Authors

• Matthew Orr, University of Utah

• Eric G. Poitras, University of Utah

• Kirsten R. Butcher, University of Utah