Empirical Study Data

该数据集汇总了来自不同实验条件下参与者的数据，旨在分析对话式解释性人工智能（XAI）界面对于用户对AI系统理解与信任的影响。数据集涵盖了五种实验条件下的参与者反馈：对照组、仪表板组、对话式XAI组、增强型对话式XAI组和大型语言模型代理组。该数据集提供了关于用户信任、对AI的理解以及对话式界面对这些因素的影响洞察。此外，数据集还包括了在不同实验条件下，诸如准确性、解释效用和用户信任分数等多种度量指标。该数据集的任务是分析使用不同解释性人工智能界面时，用户对AI系统的理解与信任程度。This dataset aggregates participant data from various experimental conditions, aiming to analyze the impact of conversational explainable artificial intelligence (XAI) interfaces on users' understanding of and trust in AI systems. The dataset covers participant feedback across five experimental conditions: control group, dashboard group, conversational XAI group, enhanced conversational XAI group, and large language model agent group. It provides insights into user trust, understanding of AI, and the influence of conversational interfaces on these factors. Additionally, the dataset includes multiple metrics such as accuracy, explanation utility, and user trust scores under different experimental conditions. The task of this dataset is to analyze users' understanding of and trust in AI systems when using different explainable AI interfaces.