Supporting Science Understanding through a Customized Learning Service for Concept Knowledge

点击翻译按钮获取中文摘要

A key educational finding from learning research is that every student brings preconceptions about how the world works to every learning situation, and that these initial understandings need to be explicitly targeted as part of an effective instructional process. This research will design and evaluate an end-to-end prototype of a customized learning service for concept knowledge (CLICK) which will enable student-centered customizations by comparing learners' conceptual understanding, depicted as concept maps, with reference domain concept maps generated from digital library resources. These comparisons will enable learning environments to provide customized retrieval, delivery, and presentation of educational resources drawn from digital libraries.This research will be conducted by a partnership between investigators at the University of Colorado at Boulder and the DLESE Program Center at UCAR. An expert study will be conducted to gather human data on reference concept map generation, comparison of student and reference concept maps, and selection of appropriate digital library resources for instructional remediation. These data will be used for developing and evaluating natural language processing algorithms that will automatically generate concept maps from DLESE resources and student work. Results from the expert study, combined with task-centered design methods, will inform the development of the CLICK service prototype. Finally, student learning will be assessed using behavioral and verbal data to address potential changes in learning processes and science understanding.This research extends current theories and techniques in adaptive learning environments. Primary outcomes of this research include the CLICK conceptual framework and the CLICK Service prototype. Together these provide a model for building effective learning environments based on educational digital libraries. Another significant outcome is the development of an automatic natural language processing-based method for identifying and representing essential science concepts that characterize deep knowledge of a targeted STEM topic.