earticle

영어

Code clones, the persistent program structures, are explored through a plenty of comprehensive techniques that sustain desirable software maintenance. Detecting the similarities of these code structures, classes, source files or directories ensure progressive development of code clones. There are adequate researches pursued in simple clones. These simple clones, recurrent in characteristics, perhaps, pave the way for the existence of much wider pattern of similarities addressed as structural clones. On the other hand, these structural clones embody large-granularity besides, the design-levels similar to program structures which are repeatedly mapped either to suit the need of designing or application of domain concept. In fact, identification of structural clones help easy comprehension of the system designing, aids software maintenance and authenticates re-engineering. The Identification and Unification of structural clones with generic program structures play a prime role in the comfortable reuse of the program. The strategy devised for this research explores the structural clones using Weighted Frequent Item-set mining (WFIM). WFIM explores the weighted frequent item-set in correspondence to a weight factor. These item-sets when identified for its recurrence in a source file or directory in same relationship are termed as structural clones. The execution of this technique discussed in this paper is performed through the two open source softwares in which the results are validated.