Prerequisite Concept Maps Extraction for AutomaticAssessment

Abstract

Traditional assessment modes usually give identical set of questions to each student, thus are inefficient for students to fix their problems. In order to perform an efficient assessment, we utilize prerequisite concept maps to find students' learning gaps and work on closing these gaps and proposed a two-phase model for concept map construction. Experiments on concept pairs with prerequisite relationships which are manually created show the promise of our proposed method. In order to meet the challenge of using concept maps in automatic assessments, we also derive a top-k concept selection algorithm which allows students to view different numbers of concepts.