Um método de detecção de plágio em códigos-fonte para disciplinas iniciais de programação

Abstract

The practice of plagiarism is a growing and serious problem in academic environment, which interferes on the quality of scientific production, affects the credibility of authoring process and threats the reputation of research institutions. This research focus on the problem of plagiarism detection in source codes developed in the initial programming classes. With the growing number of college students, the manual identification of inconsistency in the evaluations of these classes has become a practice that overcome the teachers/monitors. Despite the wide variety of available tools for plagiarism detection, only few tools are able to effectively identify plagiarism on codes developed by novice programmers, since, in most cases, these codes solve simple questions, and thus generate small solutions. Therefore, this dissertation presents a method for detecting plagiarism in activities in the initial disciplines of programming courses, in order to promote the reliability of results in the teaching-learning process. The proposed method provides satisfactory results in comparison to the methods available in the literature related to the topic.