无代码环境下的人工智能解释：用户研究

The increasing adoption of Machine Learning (ML) in sensitive areas such as healthcare, finance, and public policy has raised significant concerns regarding the transparency of automated decisions. Explainable AI (XAI) aims to address these concerns by clarifying how models generate predictions. However, most existing methods require technical expertise, limiting their accessibility for novices. This gap is particularly pronounced in no-code ML platforms, which aim to democratize AI but often lack explainability features. A human-centered XAI module has been developed and integrated into DashAI, an open-source no-code ML platform. This module incorporates three complementary techniques: Partial Dependence Plots (PDP), Permutation Feature Importance (PFI), and KernelSHAP, enhancing the workflow for tabular classification within DashAI.