Discovering places of interest using sensor data from smartphones and wearables

Abstract

Accurate detection of a user's presence at certain locations during certain times of the day is becoming increasingly important, e.g., to study the mobility, behavior, or social interaction patterns of a user and to enable the delivery of various place and time dependent services. Yet, it is often more important to determine a locale that is significant to the user, e.g., the place of work, home, homes of family and friends, social gathering places, etc., instead of geographic locations. These significant personal places can be determined through the segmentation of location traces (e.g., collected on a smartphone) into a discrete sequence of places. However, segmentation of traces with many gaps are challenging since they results in a large number of small segments, where many of these segments actually belong together. Due to recent advances in smartphone and wearable technologies, we can opportunistically obtain additional context information to determine a user's location. This work proposes a new segmentation approach that opportunistically fills gaps in location traces with the help of other types of secondary information (e.g., sleep and battery charging data), possibly from multiple devices of the same user. Using data from more than 450 users, collected over a 2-year period, we show that this approach yields fewer, but longer segments, where each segment accurately represents the presence of a user at a significant personal place.