Multi-sensor dataset of human activities in a smart home environment

Published: 26 November 2020| Version 1 | DOI: 10.17632/t9n68ykfk3.1
Gibson Chimamiwa,


This dataset is about sensor recordings of activities performed by a single user in a smart home environment. The sensors used include, passive infrared, force sensing resistors, reed switches, mini photocell light sensors, temperature and humidity, and smart plugs. The data captured include, the user's interactions with the environment, such as indoor movements, pressure applied on the bed or couch, use of the stove, TV, and fridge, or the use of electrical appliances, such as, coffee maker, dishwasher, washing machine, sandwich maker, or microwave. The dataset can be useful in several areas, including, analysis of different methods, e.g., data-driven algorithms for activity recognition or habit recognition. The dataset contains three csv files, namely, sensor, sensor_sample_int, and sensor_sample_float. The sensor file contains information, such as, the type of sensor measurement or the sensor name. The sensor_sample_int file contains sensor measurements of integer datatype while sensor_sample_float file contains sensor measurements of datatype float.



Orebro Universitet


Computer Science