ozanyurum/code-smell-representation-comparison: Initial release (v1.0)
Authors/Creators
Description
This release contains the source code used in the study:
"An Empirical Comparison of Static Software Metrics, GraphCodeBERT Embeddings, and Hybrid Representations for Software Code Smell Detection"
The repository includes the implementation for generating GraphCodeBERT embeddings; training and evaluating machine learning models using Static Software Metrics, GraphCodeBERT Embeddings, and the Hybrid Representation; performing statistical analyses; and conducting SHAP-based explainability analysis.
The SmellyCode++ dataset is not included in this release because it is publicly available at https://doi.org/10.6084/m9.figshare.28519385.v1
Files
ozanyurum/code-smell-representation-comparison-v1.0.zip
Additional details
Related works
- Is supplement to
- Software: https://github.com/ozanyurum/code-smell-representation-comparison/tree/v1.0 (URL)