UPM Institutional Repository

Automated marking system for short answer examination (AMS-SAE)


Citation

Ab Aziz, Mohd Juzaiddin and Ahmad, Fatimah and Abd Ghani, Abdul Azim and Mahmod, Ramlan (2009) Automated marking system for short answer examination (AMS-SAE). In: 2009 IEEE Symposium on Industrial Electronics and Applications (ISIEA 2009), 4-6 Oct. 2009, Best Western Premier Seri Pacific Hotel, Kuala Lumpur, Malaysia. (pp. 47-51).

Abstract

Short answers essay typed examination requires the students to write their answers in short sentences. Marking the short answers essay typed examination requires lecturers to compare the similarity of sentences from the answer scripts and marking scheme. Sentence similarity is defined as sentences that have similar meaning but they are different because of the words used or their construction structure. Automatic Marking System for Short Answers Examination (AMS-SAE) is a system that has been developed to grade students' answers based on the given marking scheme. Marks are given to the essays based on the representation of the Grammatical Relations (GRs) extracted from the sentences. The average point different between the marks computed by AMS-SAE and human are 0.049, 0.028 and 0.18. The results show that the scores are low and AMS-SAE could award similar marks as the human awarded.


Download File

[img]
Preview
Text (Abstract)
Automated marking system for short answer examination (AMS-SAE).pdf

Download (36kB) | Preview

Additional Metadata

Item Type: Conference or Workshop Item (Paper)
Divisions: Faculty of Computer Science and Information Technology
DOI Number: https://doi.org/10.1109/ISIEA.2009.5356500
Publisher: IEEE
Keywords: Malay sentences; Grammatical relations
Depositing User: Nabilah Mustapa
Date Deposited: 04 Jul 2019 04:44
Last Modified: 04 Jul 2019 04:44
Altmetrics: http://www.altmetric.com/details.php?domain=psasir.upm.edu.my&doi=10.1109/ISIEA.2009.5356500
URI: http://psasir.upm.edu.my/id/eprint/69476
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item