Online Appendix: How Effective Is Automated Trace Link Recovery in Model-Driven Development?

This folder contains 1) both non- and normalised pre-processed training data, 2) evaluation results for the different model combinations (classification x rebalancing strategy) for non-normalised training data, 3) evaluation results for the different model combinations (classification x rebalancing strategy) for normalised training data, 4) feature importance scores for the best models, and 5) source code of the application used to generate the results.