Automated marking of linear algebraic equation using n-gram method / Arsmah Ibrahim ... [et al.]

Ibrahim, Arsmah and A Bakar, Zainab and Othman, Nuru'l - 'Izzah and Ismail, Nur Fuzaina (2007) Automated marking of linear algebraic equation using n-gram method / Arsmah Ibrahim ... [et al.]. Jurnal Teknologi Maklumat dan Sains Kuantitatif, 9 (1): 1. pp. 1-14. ISSN 1823-0822

Abstract

This paper demonstrates the basic feasibility of using n-gram string matching method to grade on a set of four linear algebraic equations. The degree of correctness between the worked-out solutions submitted by respondents and the solutions schemes provided by the examiner is measured using Dice coefficient. Experiments have been conducted on 180 respondents for each question. The scores obtained by n-gram method are compared with the manually marked scores. The n-gram method is proven to be suitable when applied to these four linear algebraic equations as its performance is satisfactory and the results suggest feasibility. It is also realized that n-gram method has the advantage of producing uniformity and consistency.

Metadata

Item Type: Article
Creators:
Creators
Email / ID Num.
Ibrahim, Arsmah
arsmah@tmsk.uitm.edu.my
A Bakar, Zainab
zainab@tmsk.uitm.edu.my
Othman, Nuru'l - 'Izzah
nurul@tmsk.uitm.edu.my
Ismail, Nur Fuzaina
fuzaina@tmsk.uitm.edu.my
Subjects: Q Science > QA Mathematics > Study and teaching
Q Science > QA Mathematics > Algebra
Q Science > QA Mathematics > Evolutionary programming (Computer science). Genetic algorithms
T Technology > T Technology (General) > Information technology. Information systems
Divisions: Universiti Teknologi MARA, Shah Alam
Journal or Publication Title: Jurnal Teknologi Maklumat dan Sains Kuantitatif
UiTM Journal Collections: Others > Jurnal Akademik UiTM Sarawak
ISSN: 1823-0822
Volume: 9
Number: 1
Page Range: pp. 1-14
Keywords: N-gram, algebraic equations, dice coefficient, manual marking
Date: 2007
URI: https://ir.uitm.edu.my/id/eprint/74385
Edit Item
Edit Item

Download

[thumbnail of 74385.pdf] Text
74385.pdf

Download (1MB)

ID Number

74385

Indexing

Statistic

Statistic details