Inferring friendship network structure by using mobile phone data

TLDR

Data collected from mobile phones can provide insight into the relational dynamics of individuals. This paper compares mobile‑phone observational data with standard self‑report survey data. The authors use temporal and spatial patterns in physical proximity and calling from mobile‑phone data to infer friendships with 95% accuracy, comparing these observations to self‑report surveys. The study finds that mobile‑phone and self‑report data overlap yet differ, that friendships can be inferred with 95% accuracy from mobile‑phone patterns, and that these patterns predict individual outcomes such as job satisfaction.

Abstract

Data collected from mobile phones have the potential to provide insight into the relational dynamics of individuals. This paper compares observational data from mobile phones with standard self-report survey data. We find that the information from these two data sources is overlapping but distinct. For example, self-reports of physical proximity deviate from mobile phone records depending on the recency and salience of the interactions. We also demonstrate that it is possible to accurately infer 95% of friendships based on the observational data alone, where friend dyads demonstrate distinctive temporal and spatial patterns in their physical proximity and calling patterns. These behavioral patterns, in turn, allow the prediction of individual-level outcomes such as job satisfaction.