Dataset for the project-wise comparison of software birthmarks experiment

The dataset used for the experiment in the work titled "Project-wise Comparison of Software Birthmarks Using Weighted Partial Similarity" (https://arxiv.org/abs/2606.25418, doi: 10.48550/arXiv.2606.25418 (preprint)).Full descriptions of the dataset and experiment are provided in the manuscript. The dataset contains the following data: Extracted birthmarks (data/birthmarks); Data related to the employed projects (data/project data): dependencies, logical line count (LLC) and instruction count for each module; Comparison results (data/comparison results): top N (N = {1, 2}) highest similarity matches for each module after LLC=30 (modules with 30 or less logical lines are removed) filtering; Calculated project-wise similarities (data/project-wise similarities). Includes both unweighted and weighted results. The related Python scripts are are structured as follows: Scripts related to birthmark extraction and comparison (scripts/birthmark_management); Scripts for extraction of specific information from files (scripts/inspectors); Scripts for manipulating file/directories (scripts/managers); Scripts used for running external software (scripts/runners); Scripts related to similarity functions and project-wise similarity calculation (scripts/sim_calculation); Scripts used for threshold calculation and evaluation (scripts/thresholds); Weight calculation and assignment scripts (scripts/weight_assigner); Configuration file for various constant values (scripts/config.py). Updates to the scripts will be released on the dataset's repository, available at https://gitlab.com/ou-salab-fedorov/project-wise_comparison_of_software_birthmarks .