ADAM HARASIMOWICZ, TOMASZ DZIUBICH, ADAM BRZESKI
https://pdfs.semanticscholar.org/8d42/7d73b2746a771fd40638bd96f8f890f2b423.pdf
The presented study focused on the recognition of eight user activities (eg walking, lying, climbing stairs) basing on the measurements from an accelerometer embedded in a mobile device. It is assumed that the device is carried in a specific location of the user’s clothing. Three types of classifiers were tested on different sizes of the samples. The influence of the time window (the duration of a single trial) on selected activities and methods was investigated. A comparison with existing methods from the literature is presented.
