Automated scoring of Chinese engineering students' english essays

Liu, M; Wang, Y; Xu, W; Liu, L

Automated scoring of Chinese engineering students' english essays

Liu, M

Wang, Y Xu, W Liu, L

Permalink

Publication Type:: Journal Article
Citation:: International Journal of Distance Education Technologies, 2017, 15 (1), pp. 52 - 68
Issue Date:: 2017-01-01

Closed Access

	Filename	Description	Size
	Automated-Scoring-of-Chinese-Engineering-Students'-English-Essays.pdf	Published Version	542.34 kB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Liu, M https://orcid.org/0000-0003-4256-6531	en_US
dc.contributor.author	Wang, Y	en_US
dc.contributor.author	Xu, W	en_US
dc.contributor.author	Liu, L	en_US
dc.date.issued	2017-01-01	en_US
dc.identifier.citation	International Journal of Distance Education Technologies, 2017, 15 (1), pp. 52 - 68	en_US
dc.identifier.issn	1539-3100	en_US
dc.identifier.uri	http://hdl.handle.net/10453/126698
dc.description.abstract	© 2017 IGI Global. The number of Chinese engineering students has increased greatly since 1999. Rating the quality of these students' English essays has thus become time-consuming and challenging. This paper presents a novel automatic essay scoring algorithm called PSO-SVR, based on a machine learning algorithm, Support Vector Machine for Regression (SVR), and a computational intelligence algorithm, Particle Swarm Optimization, which optimizes the parameters of SVR kernel functions. Three groups of essays, written by chemical, electrical and computer science engineering majors respectively, were used for evaluation. The study result shows that this PSO-SVR outperforms traditional essay scoring algorithms, such as multiple linear regression, support vector machine for regression and K Nearest Neighbor algorithm. It indicates that PSO-SVR is more robust in predicting irregular datasets, because the repeated use of simple content words may result in the low score of an essay, even though the system detects higher cohesion but no spelling error.	en_US
dc.relation.ispartof	International Journal of Distance Education Technologies	en_US
dc.relation.isbasedon	10.4018/IJDET.2017010104	en_US
dc.subject.classification	Education	en_US
dc.title	Automated scoring of Chinese engineering students' english essays	en_US
dc.type	Journal Article
utslib.citation.volume	1	en_US
utslib.citation.volume	15	en_US
utslib.for	1303 Specialist Studies in Education	en_US
pubs.embargo.period	Not known	en_US
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/DVC (Teaching and Learning)
pubs.organisational-group	/University of Technology Sydney/DVC (Teaching and Learning)/Connected Intelligence Centre
utslib.copyright.status	closed_access
pubs.issue	1	en_US
pubs.publication-status	Published	en_US
pubs.volume	15	en_US

Abstract:

© 2017 IGI Global. The number of Chinese engineering students has increased greatly since 1999. Rating the quality of these students' English essays has thus become time-consuming and challenging. This paper presents a novel automatic essay scoring algorithm called PSO-SVR, based on a machine learning algorithm, Support Vector Machine for Regression (SVR), and a computational intelligence algorithm, Particle Swarm Optimization, which optimizes the parameters of SVR kernel functions. Three groups of essays, written by chemical, electrical and computer science engineering majors respectively, were used for evaluation. The study result shows that this PSO-SVR outperforms traditional essay scoring algorithms, such as multiple linear regression, support vector machine for regression and K Nearest Neighbor algorithm. It indicates that PSO-SVR is more robust in predicting irregular datasets, because the repeated use of simple content words may result in the low score of an essay, even though the system detects higher cohesion but no spelling error.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/126698