基于AST的多语言代码抄袭检测方法研究

AST-Based Multi-Language Plagiarism Detection Method

为了检测程序设计类课程中出现的作业抄袭行为,提出一种基于抽象语法树的抄袭检测方法.该方法根据多种语言的文法文件,用语法分析工具生成对应的抽象语法树(AST),利用计算生物学中序列匹配算法进行程序相似度计算,提取程序相似部分的AST特征,生成空间向量,通过聚类分析找出"抄袭团伙".实验结果表明,该方法对抄袭行为具有较好的检测效果,能准确地找出"抄袭团伙".

To detect plagiarism on programming course,AST-based plagiarism detection method is proposed.First,parsing the code generated the corresponding Abstract Syntax Tree（AST）.Biology sequence matching algorithms are used to calculate the similarity of the program.Find the similar part of the code,and then extract the AST feature in this part.Generated vector space model,and then find ＂copycluster＂ by clustered the feature.Experiments show that this method has a good effect on the detection of plagiarism and can find the ＂copy cluster＂ accurate.