Short Answer Assessment System with Student Identification using an Automatic Off-line Handwriting Recognition System and Novel Combined Features release_vuseeed425chnjeiqd5hisxt24

by Hemmaphan Suwanwiwat, University, My, Michael Blumenstein, Yongsheng Gao

Published by Griffith University.

2018  

Abstract

Examinations are a widely used form of assessment and thus important in the learning process. However, handwritten examination assessment for certain types of examination, such as essays and short answer questions, is a difficult task; it requires the markers' concentration, precision and it is time-consuming. Off-line automatic assessment systems can be an aid for teachers in the marking process. Handwriting Recognition is one of the most intensive areas of study in the field of pattern recognition. The automatic assessment of exam scripts can benefit from off-line handwriting analysis methods. There has been no recent work in the development of off-line automatic assessment systems using handwriting recognition, even though such systems will clearly benefit the education sector. The reason for this is that many schools and universities in many parts of the world still use paper-based examination. A complete off-line short answer assessment with student identification system can be an aid for teachers in the marking process as they reduce the time taken by the human marker. For the proposed system, once the automatic marking process is completed, a report on each student's mark is produced according to the name components which the system has identified. The system could reduce the time taken in marking examination papers as well as reducing the problem of mis-transcribing from examination paper to the report which may be caused by a human assessor.
In text/plain format

Archived Files and Locations

application/pdf  5.1 MB
file_drfkeodvhver7mojzujrj4qnoq
research-repository.griffith.edu.au (publisher)
web.archive.org (webarchive)
Read Archived PDF
Preserved and Accessible
Type  article-journal
Stage   published
Date   2018-01-23
Language   en ?
Work Entity
access all versions, variants, and formats of this works (eg, pre-prints)
Catalog Record
Revision: d565d892-17df-44fc-8f46-9b80ff17983d
API URL: JSON