Description of the data: The AReM dataset contains data from sensors worn on the body that are used for activity recognition. It includes readings from accelerometers, gyroscopes, and other sensors that capture human activities such as walking, standing, sitting, and other daily and sports activities. The data is segmented into time-stamped windows and labeled with the corresponding activity.
What you plan to do with the data: The intention is to use this dataset to develop and validate machine learning models capable of recognizing and classifying different human activities based on sensor data. This could have applications in health monitoring, elderly care, sports training, and human-computer interaction. The goal is to achieve a high level of accuracy in activity recognition to support the development of assistive technologies.
Information on Data provenance: The data was collected from a group of 30 volunteers who performed a series of activities while wearing body sensors. The collection process was overseen by researchers at the University of California, Irvine (UCI). The dataset was prepared and compiled by the UCI Machine Learning Repository, which is known for curating high-quality datasets for the research community. The data collection was designed to capture a wide range of motions and activities to ensure a comprehensive dataset for activity recognition research.