Subject	Signal processing
Specific subject area	Accelerometer data for Active and assisted living: technologies for extended autonomy of older people, recognition of activities of daily living (ADLs), human action recognition (HAR), de-identification of subject-dependent traits (gender, age, dominant hand).
Type of data	Table, Figure
How the data were acquired	Each subject performed the activities wearing the Empatica E4 bracelet on their dominant hand. Each subject took 30-40 minutes to complete the activities including repetitions.
Data format	Raw Filtered
Description of data collection	52 participants (33+19) were recorded in total (26 men, and 26 women). All participants were asked about their dominant hand, gender, and age. They then performed 24 different ADLs, up to 5 times each. A video (non-disclosed), was used for labelling.
Data source location	• Institution: University of Alicante (offsite data sourcing) • City/Town/Region: Alicante province (several locations) • Country: Spain
Data accessibility	Raw data is provided on the Zenodo repository at: Repository name: Zenodo (zenodo.org) Data identification number: 4750904, and 5785955 Direct URL to data: https://zenodo.org/record/4750904 (v1.0, 33 initial participants) https://zenodo.org/record/5785955 (v2.0, all 52 participants) Instructions for accessing these data: • Access the v2.0 link above, there should be three files: ADLs.csv with the names of all 24 activities (ADLs), users.csv with participant subject IDs, ages, and provided gender, and a data.zip file containing all comma-separated values (.csv) files named as described in ‘Data format’ above. Filtered data can be generated via a provided script [1].
Related research article	A. Poli, A. Muñoz-Antón, S. Spinsante, F. Florez-Revuelta, Balancing activity recognition and privacy preservation with multi-objective evolutionary algorithm, in: Proc. 7th EAI International Conference on Smart Objects and Technologies for Social Good, 2021, pp. 1–15 https://doi.org/10.1007/978-3-030-91421-9_1