XAI for CPS - Data and Trained Models

This artefact includes power traces compartmentalised into phase-specific cut files. The files are organised per experimentation scenario. They are intended to be used as the input for a data processing and Machine Learning (ML) pipeline that eventually generates SHAP scores for an XAI study. Alongside the data, three trained ML models are included, which play a role in the SHAP score calculation pipeline. These artefacts are utilised within the publication "Explainable AI to Improve Machine Learning Reliability for Industrial Cyber-Physical Systems". Refer to the provided README file for further details.本数据集包含功率轨迹数据，该类数据被划分为分阶段切片文件。所有文件均按照实验场景完成分类编排。本数据集旨在作为数据处理与机器学习（Machine Learning，ML）流水线的输入，最终为可解释人工智能（Explainable AI，XAI）研究生成SHAP评分。随数据集一同提供的还有三个经训练的机器学习模型，它们可用于SHAP评分的计算流水线。本数据集相关内容已应用于论文《可解释人工智能提升工业信息物理系统的机器学习可靠性》（"Explainable AI to Improve Machine Learning Reliability for Industrial Cyber-Physical Systems"）。如需获取更多细节，请参阅附带的README文件。