| In traditional tests, we can just only get a fraction and that can’t reflect the state knowledge of students. Cognitive diagnosis is the theory and methods that can conversion unobservable state of knowledge to response pattern that can be observed for diagnosis.In Cognitive Diagnostic we need a relations matrix between project and attributes, it indicate that the test items have what attributes and what projects to have this property (0-1Matrix). Here the attributes are the characteristics of the project of cognitive tests. One step of the Cognitive Diagnostic is to calibrate the attributes that is the calibration for project have which attributes. Also there is a certain relationship between the attributes and attributes, some cognitive diagnosis model explicitly requested the hierarchical relationship between the attributes. This article’s main job is to excavate the association between attributes.This paper combine Formal concept analysis (FCA) and the cognitive diagnostic using formal concept analysis theory method to excavate the association between attributes. The main content is in the following three steps:(1) expert calibrate a little part of the attributes, and then using an automated calibration method, calibrate attributes of the project, after the validation of the expert, and then expand to the original Q matrix.(2) verify the full Q Matrix by iterative thinking. The initial value is the original Q matrix that experts calibrated it,and the iterative method is calibration attributes and construction of concept lattices, iteration termination condition is. that the distance between the concept lattices is infinitely small.(3) Using the full Q matrix as the formal context of a concept lattice and build the concept lattice. and then dig out the correlation between the attributes.This paper generate full Q-matric by iterating, then dig hierarchical relation between attribute and’attribute. Imitation result show that this method have fine efficencice. |
PDF Full Text Request