Applications of artificial intelligence in peer assessment

Edward F. Gehringer; Ferry Pramudianto; Abhinav Medhekar; Chandrasekar Rajasekar; Zhongcan Xiao

Applications of artificial intelligence in peer assessment

Edward F. Gehringer, Ferry Pramudianto, Abhinav Medhekar, Chandrasekar Rajasekar, Zhongcan Xiao

Research output: Contribution to journal › Conference article › peer-review

1 Scopus citations

Abstract

Peer assessment has at least a 50-year history in academia, and online applications for peer assessment have been available for more than 20 years. Until recently, online applications simply transmitted classmates' feedback to each other. But in the past decade, facilities have been incorporated to automatically recognize good reviews. This helps authors know which suggestions to follow and helps reviewers improve their reviews. It can also aid in assigning peer grades. Several types of data can be used to determine review quality. These metrics can be combined using machine-learning and neural-network models to produce better estimates of review quality, and hence better estimates of the quality of reviewed work. This paper discusses past work in automatically assessing reviews, and summarizes our current efforts to build on that work.

Original language	English
Journal	ASEE Annual Conference and Exposition, Conference Proceedings
Volume	2018-June
State	Published - Jun 23 2018
Externally published	Yes
Event	125th ASEE Annual Conference and Exposition - Salt Lake City, United States Duration: Jun 23 2018 → Dec 27 2018

Keywords

Convolutional neural networks
Natural language processing
Peer assessment
Peer feedback
Peer review
Tensorflow

Cite this

@article{3528837c79044dc9856be0de09dfe0b9,

title = "Applications of artificial intelligence in peer assessment",

abstract = "Peer assessment has at least a 50-year history in academia, and online applications for peer assessment have been available for more than 20 years. Until recently, online applications simply transmitted classmates' feedback to each other. But in the past decade, facilities have been incorporated to automatically recognize good reviews. This helps authors know which suggestions to follow and helps reviewers improve their reviews. It can also aid in assigning peer grades. Several types of data can be used to determine review quality. These metrics can be combined using machine-learning and neural-network models to produce better estimates of review quality, and hence better estimates of the quality of reviewed work. This paper discusses past work in automatically assessing reviews, and summarizes our current efforts to build on that work.",

keywords = "Convolutional neural networks, Natural language processing, Peer assessment, Peer feedback, Peer review, Tensorflow",

author = "Gehringer, {Edward F.} and Ferry Pramudianto and Abhinav Medhekar and Chandrasekar Rajasekar and Zhongcan Xiao",

note = "Publisher Copyright: {\textcopyright} American Society for Engineering Education, 2018.; 125th ASEE Annual Conference and Exposition ; Conference date: 23-06-2018 Through 27-12-2018",

year = "2018",

month = jun,

day = "23",

language = "English",

volume = "2018-June",

journal = "ASEE Annual Conference and Exposition, Conference Proceedings",

issn = "2153-5965",

}

TY - JOUR

T1 - Applications of artificial intelligence in peer assessment

AU - Gehringer, Edward F.

AU - Pramudianto, Ferry

AU - Medhekar, Abhinav

AU - Rajasekar, Chandrasekar

AU - Xiao, Zhongcan

N1 - Publisher Copyright: © American Society for Engineering Education, 2018.

PY - 2018/6/23

Y1 - 2018/6/23

N2 - Peer assessment has at least a 50-year history in academia, and online applications for peer assessment have been available for more than 20 years. Until recently, online applications simply transmitted classmates' feedback to each other. But in the past decade, facilities have been incorporated to automatically recognize good reviews. This helps authors know which suggestions to follow and helps reviewers improve their reviews. It can also aid in assigning peer grades. Several types of data can be used to determine review quality. These metrics can be combined using machine-learning and neural-network models to produce better estimates of review quality, and hence better estimates of the quality of reviewed work. This paper discusses past work in automatically assessing reviews, and summarizes our current efforts to build on that work.

AB - Peer assessment has at least a 50-year history in academia, and online applications for peer assessment have been available for more than 20 years. Until recently, online applications simply transmitted classmates' feedback to each other. But in the past decade, facilities have been incorporated to automatically recognize good reviews. This helps authors know which suggestions to follow and helps reviewers improve their reviews. It can also aid in assigning peer grades. Several types of data can be used to determine review quality. These metrics can be combined using machine-learning and neural-network models to produce better estimates of review quality, and hence better estimates of the quality of reviewed work. This paper discusses past work in automatically assessing reviews, and summarizes our current efforts to build on that work.

KW - Convolutional neural networks

KW - Natural language processing

KW - Peer assessment

KW - Peer feedback

KW - Peer review

KW - Tensorflow

UR - http://www.scopus.com/inward/record.url?scp=85051223054&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85051223054

SN - 2153-5965

VL - 2018-June

JO - ASEE Annual Conference and Exposition, Conference Proceedings

JF - ASEE Annual Conference and Exposition, Conference Proceedings

T2 - 125th ASEE Annual Conference and Exposition

Y2 - 23 June 2018 through 27 December 2018

ER -

Applications of artificial intelligence in peer assessment

Abstract

Keywords

Other files and links

Cite this