The real dataset used in our AISTATS'17 work contains the data for 5 different types of measurements from 50 rooms in an office building.

Each room includes 5 types of measurements: CO2 concentration, room air humidity, room temperature, luminosity, and PIR motion sensor data.

Each file contains the time stamps (in Unix Epoch Time) and acutal readings from each sensor.

The data set is available here.


If you use the dataset, please consider citing the following paper:

Dezhi Hong, Quanquan Gu, Kamin Whitehouse.
High-dimensional Time Series Clustering via Cross-Predictability.
In AISTATS'17 (bibtex)