PERCY
收藏资源简介:
PERCY数据集是由悉尼科技大学开发的,旨在捕捉多模态信息的人机交互对话数据集。该数据集包含30名参与者的真实交互数据,涵盖了个人兴趣、情感状态等多方面的信息。数据集的创建过程包括参与者填写问卷、与机器人进行对话,并利用GPT-4生成基于参与者情感状态和个人信息的相关回应。PERCY数据集的应用领域广泛,主要用于训练能够提供个性化和情感感知的人机交互系统,旨在解决传统对话系统在深度交互和情感理解方面的不足。
The PERCY Dataset is a human-computer interaction dialogue dataset developed by the University of Technology Sydney for capturing multimodal information. This dataset contains real interaction data from 30 participants, covering diverse information such as personal interests and emotional states. The dataset creation process involves participants completing questionnaires, engaging in conversations with robots, and using GPT-4 to generate relevant responses tailored to the participants' emotional states and personal information. The PERCY Dataset has a wide range of application scenarios, mainly used for training human-computer interaction systems that can provide personalized and emotion-aware services, aiming to address the shortcomings of traditional dialogue systems in deep interaction and emotional understanding.




