Hybrid approach for Detecting Code Clone by Metric and Token based comparison

Deepali, Deepali, Ankur Gupta, Chirag Batra


In software development process, coping of
existing code fragment and pasting them with or without
modification is a frequent process. Clone means copy of an
original form or duplicate. Software clone detection is
important to reduce the software maintenance cost and to
recognize the software system in a better way. There are
many software code clone detection techniques such as textbased,
token-based, Abstract Syntax tree based etc. and they
are used to spot and finding the existence of clones in
software system. One of the approaches to detect code clones
is by analysis of different metrics for the programs. Another
approach is token based comparisons of two programs to
detect code clone. Our technique uses the combination of
metrics and token based approach using hashing algorithm
by analysis of two source codes in the process of finding
clones among them and the percentage of cloning is reported
as result.

Keywords— software clone detection, metric based
comparison, token based approach using hashing algorithm,
java byte code.

Full Text:


DOI: https://doi.org/10.26483/ijarcs.v7i6.2810


  • There are currently no refbacks.

Copyright (c) 2016 International Journal of Advanced Research in Computer Science