Code Clone Detection: A Systematic Review

Abstract

Code cloning in software systems has gained significant development in past few years. Cloning is a general mean of reusing software as existing code snippets can be utilized either by copy and paste methods or by minor modifications in the current code in software systems. However, this may lead to produce bugs and maintenance issues. A plethora of various code clone detection
tools and techniques have emerged from last few decades. However, there are no comprehensive studies reviewing all the available techniques since 2013. The aim of this Systematic Literature Review (SLR) is to fill this research gap by systematically reviewing all the available research and extending the research on this particular topic. The main objectives of the study are to
identify, categorize and synthesize relevant techniques related to this particular topic. After analyzing initial set of 1181 studies gathered from four large databases, 37 studies relevant to defined research questions were identified by following a systematic and unbiased selection procedure according to standard PRISMA guidelines. This selection process is followed by the data extraction, detailed analysis and reporting of findings. The results of this SLR reveals that different tools and techniques have widely been used for code clone detection, but graph-based and metric-based approaches are most prolific approaches. These approaches have also been used as a part of hybrid approaches. Different match detection techniques are also reported. However, to cope with rapidly evolving clones in software systems, the need is to develop more efficient techniques to improve the state of current research. This study concludes with new recommendations for future research.