.png){kind=icon}
.png){kind=icon}
Return
Code clone detection based on semantic images
.png){kind=icon}
PRE
AI
DOI：10.1016/j.infsof.2026.108155.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Proposes semantic image transformation by converting code ASTs into fixed-size grayscale images, reducing structural complexity and computational cost. • Integration of community detection, centrality analysis, and statement sequence embedding to capture both structural and semantic features of code. • Achieves high efficiency and accuracy for Type-4 (semantic) clone detection, outperforming 8 state-of-the-art methods on BigCloneBench and Google Code Jam datasets with significantly reduced computational time.
Keywords:
semantic image transformation
code clone detection
community detection
centrality analysis
statement sequence embedding
Journal
IF:
4.3
Papers:
3.7K
Citations:
7.7K