Penerapan Algoritma Levenshtein Distance pada Aplikasi Asisten Guru Bahasa Inggris

Abstract

Many plagiarism acts among high school students when submitting assignments and lacking systems which are able to detect similarities in the contents of students’ tasks. In a nutshell, the teacher finds it difficult to get the authenticity of the tasks done by students. This study aims at creating an English teacher assistant application, to make it easier to correct the authenticity of students’ essay assignments so that the teacher can get the task that is the original work of the students themselves without any act of plagiarism. This application is divided into two stages, namely the stages of the text preprocessing and then the similarity calculation is done with the Levenshtein Distance algorithm. Stages of text preprocessing consist of tokenizing, purifying, stopword removal, stemming, and sorting. To get text similarity scores between students, calculations are made with the Levenshtein Distance algorithm by applying the limit of the text similarity value (threshold) to 70%, if the students’ text similarity is less than 70%, the document will be received by the system and entered into the database. Text similarity exceeds 70%, then the document will be rejected. Therefore, the results of this study can help English teachers, especially educational institutions, to obtain original documents and reduce plagiarism from an early age at the high school level.