An automatic off-line short answer assessment system using novel hybrid features

Suwanwiwat, Hemmaphan, Pal, Umpanda, and Blumenstein, Michael (2016) An automatic off-line short answer assessment system using novel hybrid features. In: Proceedings of the International Conference on Digital Image Computing: Techniques and Applications. pp. 757-764. From: DICTA 2016: International Conference on Digital Image Computing: Techniques and Applications, 30 November - 2 December 2016, Gold Coast, QLD, Australia.

[img] PDF (Published Version) - Published Version
Restricted to Repository staff only

View at Publisher Website: http://dx.doi.org/10.1109/DICTA.2016.779...
 
4


Abstract

To date, paper-based examinations are still in use worldwide on all levels of education levels (e.g. secondary, tertiary levels). However, literature regarding off-line automatic assessment systems employing off-line handwriting recognition is not numerous. This paper proposes an off-line automatic assessment system employing a hybrid feature extraction technique - a newly proposed Modified Direction and Gaussian Grid Feature (MDGGF), along with its enhanced technique. In this study other original feature extraction techniques, together with their enhanced features, were also used for feature extraction technique efficiency comparison purposes. Classifiers, namely artificial neural networks and support vector machines, were selected to be employed in the experiments. Two types of datasets were employed in the experiment for both feature extraction technique accuracy and efficiency comparisons. The best correctly recognised rate of 98.33% with 100% accuracy was obtained when employing the proposed MDGGF to the off-line automatic assessment system.

Item ID: 49109
Item Type: Conference Item (Research - E1)
ISBN: 978-1-5090-2896-2
Keywords: artificial neural networks, Gaussian grid feature, handwriting recognition, modified direction feature, off-line automatic assessment system
Date Deposited: 21 Jun 2017 02:42
FoR Codes: 46 INFORMATION AND COMPUTING SCIENCES > 4603 Computer vision and multimedia computation > 460306 Image processing @ 20%
46 INFORMATION AND COMPUTING SCIENCES > 4611 Machine learning > 461199 Machine learning not elsewhere classified @ 80%
SEO Codes: 89 INFORMATION AND COMMUNICATION SERVICES > 8999 Other Information and Communication Services > 899999 Information and Communication Services not elsewhere classified @ 100%
Downloads: Total: 4
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page