Continuous Ambient-Sensor Activity Data from Smart Homes
Catalog Summary
"Continuous smart-home sensor events and activity-recognition representations; UCI reports 13,956,534 instances, with raw and engineered-feature files."
Catalog Notes
Overview
This CASAS-related collection supports activity recognition from continuous ambient sensing in smart homes. The source describes motion, door, temperature and light sensors with activity annotations.
Representations
The repository distributes raw event files and derived feature representations. Its raw-event description lists Date, Time, Sensor, Translate01, Translate02, Message, SensorType and Activity fields; engineered representations use sliding windows of 30 sensor events. Do not assume that all distributed files share a single schema or that raw and derived rows are independent observations.
Verification limits
UCI reports 13,956,534 instances and missing values. The approximately 13 GB archive was not downloaded or fully recounted in this audit; this record’s row_count retains the publisher-reported figure. File-specific documentation and missing-value handling must be checked before training. The citation uses the source’s 2012 citation year, distinct from its 2019 UCI donation date.
Source and provenance
Dataset DOI: https://doi.org/10.24432/C5D60P.
View Data Structure
To explore column names, data types, and sample rows, visit the official dataset page on UCI Machine Learning Repository.
Preview on UCI Machine Learning RepositoryCite This Dataset
Cook, D., Crandall, A., & Thomas, B. (2012). Human Activity Recognition from Continuous Ambient Sensor Data [Data set]. UCI Machine Learning Repository. https://doi.org/10.24432/C5D60P
Source metadata: UCI Machine Learning Repository (2012) · DOI: 10.24432/C5D60P
Indexed by IoTDataset.com on Sep 14, 2026
Review the Source Record
Confirm the licence, version, access conditions, file format, and provenance at the source before use.