Older Adult Screening Interaction and Support Dialogue Dataset

This dataset was collected from a human–assistant interaction study in the context of health screening for older adults. The study aims to investigate the usability, interaction experience, and task completion of voice assistants during cognitive and health screening tasks. During data collection, participants interacted with different types of voice assistant systems and completed corresponding screening tasks. The dataset can be used for research on human–computer interaction, voice assistants, health screening for older adults, and dialogue system evaluation. The publicly available data in this repository/platform only include de-identified materials that are suitable for public sharing, including but not limited to: De-identified human–assistant dialogue transcripts; Structured annotations such as dialogue turns, task types, and system types; No directly identifiable personal information, raw facial images, raw videos, raw speech recordings, or other sensitive biometric information is included. In the publicly released data, we have made efforts to remove or replace participant names, contact information, specific addresses, institution names, identification information, and other content that may lead to personal identification. As this study involves health screening for older adults, some raw data contain sensitive information, including participants’ facial images, videos, voices, speech recordings, and health-related expressions. Such information may involve risks related to biometric identification and privacy protection. Therefore, the following data are not included in the publicly released dataset: Raw speech recordings; Raw video data; Facial images or video frames containing facial information; Raw records that could be used to identify participants; Other data that may involve personal privacy, health information, or biometric information. Due to ethical review requirements, informed consent restrictions, and privacy protection considerations, the above sensitive data cannot be made publicly available for direct download. Researchers with a legitimate academic need for non-public data may contact the authors by email and submit a data access request. The request should include: The applicant’s name, affiliation, and contact information; The research purpose and specific use case; The type of data requested; Data security storage and privacy protection measures; Whether the data will be redistributed, publicly released, or used for commercial purposes. Requests will be evaluated based on ethical approval requirements, the intended use of the data, the scope of participants’ informed consent, and privacy protection requirements. Restricted access may only be granted if the relevant conditions are met and a necessary data use agreement is signed. Contact email: [e-mail]