Software Clone Detection Basedon Context Information

Software clone detection is very promising and innovative within the industryfield. Existing mainstream clone detection techniques mainly focus ondetecting the similarity of source code itself, which makes them capable ofdetecting Type I and Type II clones (Type I clones are two identical codefragments except for variations in format and Type II clones are twostructurally identical code fragments except for variations in format). Butthey rarely pay attention to the relationship between codes. It becomes animportant research area to detect Type III code clones, which are clones withminor difference in statements, by using the context information in thesource code. I carry out a detailed analysis of existing software clone detectiontechniques in this thesis. It raises issues of existing software clone detectiontechniques in theory and practice. On the basis of the analysis, I propose anew method to improve existing clone detection techniques with a detailedtheory analysis and experimental verification. This method makes detectionof Type III software clones possible