Tree Slicing in Clone Detection: Syntactic Analysis Made (Semi)-Semantic

Nowadays most of software contains code duplication that leads to serious problems in software maintenance. A lot of different clone detection approaches have been proposed over the years to deal with this problem, but almost all of them do not consider semantic properties of the source code. We propose to reinforce traditional tree-based clone detection algorithms by using additional information about variable slices. This allows to find intertwined/gapped clones on variables; preliminary evaluation confirms applicability of our approach to real-world software.