Model-based cognitive diagnosis of students' test performance in an e-learning environment

Authors:
Rong Chen;Junjie Xu;Yingjie Song;Wu Deng;Yanheng Li
Affiliations:
College of Information Science and Technology, Dalian Maritime University, Dalian, China;College of Information Science and Technology, Dalian Maritime University, Dalian, China;College of Information Science and Technology, Dalian Maritime University, Dalian, China;College of Information Science and Technology, Dalian Maritime University, Dalian, China;College of Information Science and Technology, Dalian Maritime University, Dalian, China
Venue:
ICWL'10 Proceedings of the 2010 international conference on New horizons in web-based learning
Year:
2010

Citing 3
Cited 0

A theory of diagnosis from first principles

Artificial Intelligence
DiKe - A Model-Based Diagnosis Kernel and Its Application

KI '01 Proceedings of the Joint German/Austrian Conference on AI: Advances in Artificial Intelligence
Diagnosing program errors with light-weighted specifications

IEA/AIE'06 Proceedings of the 19th international conference on Advances in Applied Artificial Intelligence: industrial, Engineering and Other Applications of Applied Intelligent Systems

Quantified Score

Hi-index	0.01

Visualization

Abstract

Cognitive diagnosis is the process of inferring a cognitive state from observations of performance. This paper considers the problem of cognitive diagnosis as an instance of model-based diagnosis, as studied in artificial intelligence for many years. The model-based cognitive diagnosis we present runs on a model of students' courses in terms of knowledge items that they may learn, tests them and helps them to understand their faults in cognition, and thus improve their learning performance in an E-learning environment. To do so, courses are formally defined as set of knowledge items with requirement constraints, knowledge items are associated with a set of exam questions. Moreover, diagnostic algorithms are used to help a student understand what knowledge item within a course the student does not master, the root reason of his/her test errors, and the recommendations like what should be done next. Experimental results show that the group of students with such understanding can improve their testing performance greatly in an E-learning environment.