I think it's inherently interesting to study problem-solving sequences and methodologies (towards a theory of scientific growth, as Larry Laudan would say). This article addresses what it calls a "drawback" of existing sequence-based approaches, "the absence of a human in the loop" because "existing techniques often do not present sequences in an interpretable manner, either due to lack of visualization or because of the scale of the data." This study looks at student problem-solving in a learning game called Parallel. Student sequences are compared with each other and with what might be called the 'expert trace', "the sequence an expert would generate given the educational environment." This is to me an example of what we used to call automated competency detection and recognition (ACDR), an approach that ultimately went unexplored in our LPSS project of seven years ago.
Today: 5 Total: 101 [Direct link] [Share]
