This repository contains the data and code used for the analysis of the misuse of the BigCloneBench dataset in the paper "How the Misuse of a Dataset Harmed Semantic Clone Detection".
We provide the following data as supplementary material:
- The subset of 406 clone pairs used for the manual investigation.
- The protocol for the manual investigation.
- The results of the manual investigation.
- The analysis results of the LLM-based evaluation of the dataset.
- The code used for the LLM-based analysis of the literature review.
- The results of the literature review.
The data is separated into two folders:
BCB406: Contains the data related to the 406 clone pairs used for the manual investigation.Literature: Contains the data related to the literature review.
The folder IWSC2022 contains the data related to a preliminary version of the analysis, which appeared in the 16th International Workshop on Software Clones (IWSC 2022).