A development of an integrated programming class support system focusing on the similarity based on the content of source codes and author's coding style

2014 Fiscal Year Final Research Report

• Project/Area Number: 24700918
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Educational technology
• Research Institution: Osaka Sangyo University
• Principal Investigator: OHNO Asako 大阪産業大学, 工学部, 講師 (90550369)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: 知的学習システム / 教育工学 / 授業支援システム / 盗用発見 / 類似性検出 / 採点支援 / 授業課題ソースコード / 特徴抽出

Outline of Final Research Achievements

In this study, I improved the two source code similarity measuring methods that I had proposed in my former study, that were, "A similarity measuring method using reference vector (FRef)" and "A similarity measuring method based on author’s coding style (CM Algorithm)". I implemented the two methods to a programming class support system to achieve auto-scoring, plagiarism detection, and coding style evaluation functions. From the results of the evaluation experiments using a set of source codes produced by students in real-world programming class as test data, I confirmed high performance for plagiarism detection and coding style evaluation function while scoring function needed more improvement to be used in practice.

Free Research Field

知的学習システム