PsyInterview
收藏资源简介:
PsyInterview数据集由香港大学和清华大学的研究团队创建,旨在通过合成临床访谈数据来支持情感障碍的筛查。该数据集包含1157条多轮访谈对话,涵盖了情感障碍、其他精神障碍及健康对照组。数据来源广泛,包括临床案例手册、研究论文和开源数据集。每条对话平均包含14个话语,每个话语平均24个词。数据集通过四阶段的数据生成管道创建,确保了数据的多样性和真实性。PsyInterview数据集的应用领域主要集中在心理健康领域,特别是情感障碍的筛查和临床访谈自动化,旨在减轻心理医生的工作负担并提高筛查效率。
The PsyInterview dataset was developed by a research team from The University of Hong Kong and Tsinghua University, aiming to support the screening of affective disorders via synthetic clinical interview data. This dataset contains 1,157 multi-turn interview conversations, covering three groups: patients with affective disorders, those with other mental disorders, and healthy controls. It draws data from diverse sources, including clinical case manuals, research papers, and open-source datasets. Each conversation contains an average of 14 utterances, with each utterance averaging 24 words. The dataset was constructed via a four-stage data generation pipeline, ensuring its diversity and authenticity. The PsyInterview dataset is primarily applied in the field of mental health, particularly for affective disorder screening and clinical interview automation, with the goal of reducing the workload of psychologists and improving screening efficiency.




