Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.
Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Case-based cognitive modeling: A student modeling methodology for an intelligent tutoring system
Abstract
In contrast with conventional computer-aided instruction (CAI) systems, intelligent tutoring systems (ITS) can provide adaptive instruction for individual students by diagnosing the student's current understanding during a tutoring session and fitting the instruction according to the student model. The student modeling task can be considered a design problem which involves a process for constructing and manipulating a data structure representing the student's current understanding of the domain knowledge being delivered. Focusing on the student diagnosis component, this dissertation investigates whether or not the case-based reasoning (CBR) technique can be utilized to improve the effectiveness and efficiency of diagnosis ability for an intelligent tutoring system. Case-based diagnosis (CBD), unlike prevailing rule-based reasoning approaches, does not require a complete library of every possible faulty behavior at the beginning. Instead, highlighting experience as the central feature of diagnostic expertise, CBD starts with a minimal set of unsatisfactory situations and increases its diagnostic capability by adapting the prior cases (experiences) to new situations. This study explores case-based student diagnosis issues, such as choices of indexes to be used for organizing prior cases, methods for choosing the most relevant cases, and general formulations of adaptation heuristics used to modify previous cases to fit the new case. For this study, a cognitive diagnosis system, called Case-Based Cognitive Modeling (CBCM) system, has been implemented in Common LISP. The system has been experimentally evaluated by applying to the goal programming model formulation task. It has been found that once the system trained with enough number of cases, the system's diagnosing performance is comparable to that of human tutors.
Subject Area
Educational software|Management|Artificial intelligence|Business education
Recommended Citation
Kim, Joong Han, "Case-based cognitive modeling: A student modeling methodology for an intelligent tutoring system" (1993). ETD collection for University of Nebraska-Lincoln. AAI9415973.
https://digitalcommons.unl.edu/dissertations/AAI9415973