Spelling Correction Application with Damerau-Levenshtein Distance to Help Teachers Examine Typographical Error in Exam Test Scripts

¹ Faculty of Information Technology, Tarumanagara University, Jl. Letjen. S. Parman no. 1, Jakarta 11440, Indonesia
² School of Computing, National University of Singapore, 13 Computing Drive, 117417 Singapore

^* Corresponding author: viny@untar.ac.id

Abstract

This research was intended to create Spelling Correction Application to help teachers examine questions scripts with the capability to found typographical error and give suggestion for non-real word error. This application is built with simple Damerau-Levenshtein Distance method to detect errors and give word suggestions from the typo word. This application can be used by the teacher to examine documents in the form of short answer, essay and multiple choices then save them back in the form of original documents. This application is built using a dictionary lookup consist of 41 312 words in Indonesian. The first test result is the application can detect non-real word errors from 50 sentences that have non-real word error in each sentence and produce an accuracy of 88 %. The second test is try to detect typographical error in exam test script that consist of 15 sample questions, consisting of five essay questions, five short answer, and five multiple choices.