期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2012
卷号：1
期号：8
页码：298-301
出版社：Shri Pannalal Research Institute of Technolgy
摘要：Copy-pasted code is very common in largesoftware and product line software becauseprogrammers prefer reusing code via copy-paste inorder to reduce programming effort. Copy pasted codeis prone to introducing errors. Unfortunately, it ischallenging to efficiently identify copy-pasted code inlarge software. Existing copy-paste detection tools areeither not scalable to large software, or cannot handlesmall modifications in copy-pasted code. In this paperwe propose an enhanced CloSpan algorithm for CPMinertool that uses data mining techniques toefficiently identify copy-pasted code in large softwareincluding operating systems. Specifically, it takes lessthan 20 minutes for CP-Miner with enhanced CloSpanAlgorithm to identify 190,000 copy pasted segments inLinux and 150,000 in FreeBSD.
关键词：software product lines; code reuse; code;duplication; data mining.
