We present a study to detect friendship, its strength, and its change from smartphone location data collectedamong members of a fraternity. We extract a rich set of co-location features and build classifiers that detectfriendships and close friendship at 30% above a random baseline. We design cross-validation schema to testour model performance in specific application settings, finding it robust to seeing new dyads and to temporalvariance.
@article{arxiv.2008.02919,
title = {Can Smartphone Co-locations Detect Friendship? It Depends How You Model It},
author = {Momin M. Malik and Afsaneh Doryab and Michael Merrill and Jürgen Pfeffer and Anind K. Dey},
journal= {arXiv preprint arXiv:2008.02919},
year = {2020}
}