Multimodal Online Federated Learning Version UCI-HAR dataset
收藏IEEE2026-04-17 收录
下载链接：
https://ieee-dataport.org/documents/multimodal-online-federated-learning-version-uci-har-dataset
下载链接
链接失效反馈官方服务：
资源简介：
The UCI-HAR dataset is a widely recognized resource for human activity recognition research. It includes 10299 data samples collected from 30 participants (average age: 24) engaging in six activities (six classes): walking, walking upstairs, walking downstairs, sitting, standing, and lying down. These activities were recorded using smartphone sensors, specifically accelerometers and gyroscopes, which capture three-dimensional motion data. The sensors sampled data at 50 Hz, producing 128 readings per sensor axis within each time window. This dataset is utilized in our experiments to analyze sensor-based human activity recognition using three-dimensional motion data.
提供机构：
Heqiang Wang
