The design of a student modeling system for binary operations

Jeffrey Chua

Date of Publication

1998

Document Type

Master's Thesis

Degree Name

Master of Science in Computer Science

Subject Categories

Computer Sciences

College

College of Computer Studies

Department/Unit

Computer Science

Thesis Adviser

Rhodora Reyes

Defense Panel Chair

Jose Lloyd Espiritu

Defense Panel Member

Kai Shan Fernandez
Raymund Sison

Abstract/Summary

This study designs a student model for Binary operations using the constraint-based approach. Constraint-based modeling is a student modeling approach that focuses on student errors. Domain knowledge is represented as procedural knowledge, which serves as constraints and is used to identify the errors. The constraint-based student model for binary operations presented supports procedural learning. The student can learn about concepts and procedures in solving binary operations. A total of 23 constraints were used in solving binary operations. Seven constraints were used for binary addition, 8 were used for binary subtraction, 4 for binary multiplication and 4 for binary division. Each of these constraints were based from the procedures involved in solving binary operations. This design was chosen because of its potential advantages. First, there is no need for large scale empirical studies of domain experts (although interviews with experts might be helpful in identifying key ideas in a domain) and there is no need for large scale empirical studies of students to create and validate the bug library. Also, there is no need to create computationally expensive AI inference mechanisms. Constraints can be compared to problem states with off-the-shelf pattern matching algorithms. Finally, the constraint-based student model technique is neutral with respect to teaching. A student can be tutored after each constraint violation or after a sequence of them as seems most appropriate for the domain.

Constraint-based model, on the other hand, has some potential disadvantages. For some domains, it might be impossible to identify properties of problem states which are highly informative with respect to the student's understanding. However, for procedural types of problem (e.g., binary operations) constraint-based model is an ideal approach to student modeling. There are many possibilities in extending this research. The evaluation study will provide the data for completing the entire intelligent tutoring system which includes the tutor model. The tutor model will be responsible for presenting the lessons, exercises, examples and the correction or remediation to the errors determined by the constraint-based student model. Focus on the development of the set of teaching rules that would govern the selection of appropriate amount of help for each student is highly recommended. Based from the evaluation of the student model design, the percentage shows that the design recognizes all types of errors students commit every time the constraints are violated. Future implementation of this design requires validation of the evaluation of the student model by getting user feedback from the student as well as the teachers who use the system. The evaluator can then check the percentage of correct evaluation of the student model.

Abstract Format

html

Language

English

Format

Print

Accession Number

TG02830

Shelf Location

Archives, The Learning Commons, 12F Henry Sy Sr. Hall

Physical Description

49, [34] leaves; 28 cm.

Keywords

Intelligent tutoring systems; Binary-coded decimal system; Coding theory; Computer arithmetic and logic units

This document is currently not available here.

Share

COinS