Human activity modeling from large-scale sensor data is an emerging domain. We present a framework to classify days into two groups: weekends and weekdays. The data collected by Device Analyzer, an Android application managed by University of Cambridge, includes cell tower connectivity data, from which physical location can be derived. Since the location information is removed from the datasets, the semantic of places needs to be inferred from the connectivity patterns. In this particular experiment, we use cell tower data to identify weekends and weekdays. By processing data collected over a long period of time by Device Analyzer, we find the cell towers which are mainly used in weekends or weekdays and then take advantage of them to identify the day type.
@inproceedings{DBLP:conf/mum/SadriS14,
author = {Amin Sadri and
Flora Dilys Salim},
bibsource = {dblp computer science bibliography, https://dblp.org},
biburl = {https://dblp.org/rec/conf/mum/SadriS14.bib},
booktitle = {Proceedings of the 13th International Conference on Mobile and Ubiquitous
Multimedia, Melbourne, VIC, Australia, November 25-28, 2014},
doi = {10.1145/2677972.2678006},
editor = {Arkady B. Zaslavsky and
Seng W. Loke and
Lars Kulik and
Evaggelia Pitoura},
pages = {244--247},
publisher = {ACM},
timestamp = {Mon, 16 Sep 2019 01:00:00 +0200},
title = {Day type classification using cell tower connectivity data from smartphones},
url = {https://doi.org/10.1145/2677972.2678006},
year = {2014}
}
© 2021 Flora Salim - CRUISE Research Group.