No abstract

No abstract

No abstract

No abstract

One way to optimize learning processes is to clearly inform the learner about problematic areas. Recent work on gaze-based CHI showed that a reader's language skill can be inferred by gaze analysis. However, only few approaches have been proposed to identify those document parts a reader finds problematic. Our goal is to develop a computational method for reading incomprehension extraction. As initial work, we analyze which eye gaze features are useful for such part-based reading incomprehension extraction at three levels of document structure: paragraphs, segments and words.