Source Code Clone Detection Using Unsupervised Similarity Measures
-
Updated
Dec 17, 2025 - Java
Source Code Clone Detection Using Unsupervised Similarity Measures
A multilingual syntatic code clone detector based on ANTLR parser generation.
FortranAS is an [Antlr4 🔗](https://github.com/antlr/antlr4) powered FORTRAN parsing and code clone detection tool. FortranAS leverages [grammars-v4 🔗](https://github.com/antlr/grammars-v4/tree/master/fortran) to generate parsers and lexers and supports any FORTRAN version with available corresponding grammars.
Research project on the automatic classification of code clones according to type, as well as an attempt to filter out False Positive clones.
Token-based code clone detection combining 16 unsupervised similarity measures with XGBoost. 95.9% F1 on BigCloneBench, 82.2% on Google Code Jam.
To associate your repository with the code-clone-detection topic, visit your repo's landing page and select "manage topics."