Can Smartphone Co-locations Detect Friendship? It Depends How You Model It
This work addresses the challenge of inferring social relationships from co-location data, but it is incremental as it applies existing methods to a specific domain with new features.
The study tackled the problem of detecting friendship, its strength, and changes using smartphone location data from a fraternity, achieving a 30% improvement over a random baseline in classification accuracy.
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.