未明确命名

本研究旨在通过开发一种偏好诱导界面，收集包含丰富用户信息的密集推荐系统数据集。该数据集将涵盖从照片、绘画到屏保设计等多个视觉体验领域，通过复制Jester界面的设计并增加新元素来收集数据。数据集的创建过程中，用户首先接受界面使用指导，随后在随机展示的视觉刺激前提供人口统计信息，并通过滑动条进行项目评估。此数据集旨在促进推荐系统研究与社会行为科学的融合，用于评估人类偏好的多样性，测量心理效应，并支持反事实学习实验的评估。

This study aims to develop a preference-eliciting interface to collect a dense recommender system dataset with rich user information. This dataset spans multiple visual experience domains, ranging from photographs, paintings to screensaver designs, with data collected by replicating the Jester interface design and adding new elements. During the dataset creation workflow, users first receive guidance on using the interface, then provide demographic information while viewing randomly presented visual stimuli, and evaluate the items via sliders. This dataset is designed to facilitate the integration of recommender system research and social behavioral science, enabling the assessment of human preference diversity, measurement of psychological effects, and evaluation of counterfactual learning experiments.