Student Paper Comparison System using Kolmogorov Complexity and Diff Algorithm

Abstract

In the academe, students sometimes rely on copying each other’s works or previous works in order to meet
requirements. This action leads to plagiarism, which is becoming part of academic institutions concern in reducing
growing academic dishonesty. Though commercialized plagiarism checkers are available, some are not designed to
compare submitted papers while some only allows to compare two up to documents. With regards to such issue, this
study aims to design and develop a comparison system for students’ papers that is capable of uploading multiple
documents and calculating the similarity between the documents. Moreover, the application was created using HTML,
CSS, JavaScript, PHP, and MySQL. The developed system consists of two main modules: The Document Upload which
allows users to upload multiple documents for comparison and manage the stored documents, and the Document
Comparison, which serves as the mechanism of the system for detecting possible plagiarism. An algorithm based on
Kolmogorov Complexity was utilized to measure the similarity between the documents while the Diff Algorithm was
used to highlight the parts of the document that is suspected to be plagiarized. Furthermore, tests were performed to
verify if the system is functioning as expected and to measure the accuracy of the output from the system. Eighty-one
respondents evaluated the developed system using the ISO 25010 software quality model in terms of the Product
Quality. An overall mean of 4.63 which can be interpreted as “Excellent” in descriptive terms validates that the
objectives of the study were met and achieved, and further indicates that the system was developed according to its
desired functions and requirements.