Supplementary Material (code and data sets from Nokia under CC BY NC ND 4.0 license) for the following paper: L. Madeyski and S. Stradowski, “Predicting test failures induced by software defects: A lightweight alternative to software defect prediction and its industrial application,” Journal of Systems and Software, p. 112360, 2025. DOI: 10.1016/j.jss.2025.112360 URL: https://doi.org/10.1016/j.jss.2025.112360|软件缺陷预测数据集|测试失败预测数据集

This package (see also https://madeyski.e-informatyka.pl/download/MadeyskiStradowski24Supplement.pdf) includes research artefacts (developed code and datasets from Nokia under CC BY NC ND 4.0 license) required to reproduce the results presented in the paper:Lech Madeyski and Szymon Stradowski, “Predicting test failures induced by software defects: A lightweight alternative to software defect prediction and its industrial application,” Journal of Systems and Software, p. 112360, 2025. DOI: 10.1016/j.jss.2025.112360 URL: https://doi.org/10.1016/j.jss.2025.112360 Highlights from the paper:We propose a Lightweight Alternative to Software Defect Prediction (LA2SDP)The idea behind LA2SDP is to predict test failures induced by software defectsWe use eXplainable AI to give feedback to stakeholders \&amp; initiate improvement actionsWe validate our proposed approach in a real-world Nokia 5G test processOur results show that LA2SDP is feasible in vivo using data available in Nokia 5G<br>