LifeSnaps, a 4-month multi-modal dataset capturing unobtrusive snapshots of our lives in the wild

Abstract

Ubiquitous self-tracking technologies have penetrated various aspects of our lives, from physical and mental health monitoring to fitness and entertainment. Yet, limited data exist on the association between in the wild large-scale physical activity patterns, sleep, stress, and overall health, and behavioral and psychological patterns due to challenges in collecting and releasing such datasets, including waning user engagement or privacy considerations. In this paper, we present the LifeSnaps dataset, a multi-modal, longitudinal, and geographically-distributed dataset containing a plethora of anthropological data, collected unobtrusively for the total course of more than 4 months by n = 71 participants. LifeSnaps contains more than 35 different data types from second to daily granularity, totaling more than 71 M rows of data. The participants contributed their data through validated surveys, ecological momentary assessments, and a Fitbit Sense smartwatch and consented to make these data available to empower future research. We envision that releasing this large-scale dataset of multi-modal real-world data will open novel research opportunities and potential applications in multiple disciplines.

Examining Participant Adherence with Wearables in an In-the-Wild Setting

“…Other similar in-the-wild studies with wearable technology have achieved similar compliance rates, which further demonstrates the feasibility of experiments with comparable study designs. One study on the Fitbit reported a high user engagement for 73% of the total days of the experiment [ 12 ], although it did not account for daily adherence to wearing the device and consistency of usage. In another in-the-wild study that also made use of the Oura Ring to examine the influence of maintaining a sleep tracking routine in one’s daily life, users who were noted to rarely check their Oura application still wore the ring 85% of nights throughout the experiment duration.…”

“…There is one experiment that was conducted in-the-wild with the Fitbit that ran with a decently sized generalized sample over an extended period of time. However, they do not discuss granular adherence to the device at an individual level and only provide aggregate information on how often the Fitbit was worn [ 12 ].…”

UnStressMe: Explainable Stress Analytics and Self-tracking Data Visualizations

“…Our models are developed on the LifeSnaps dataset [9], a real-world, multi-modal dataset containing 71M rows of data collected from 71 users, in four countries, from May '21 to Feb. '22. LifeSnaps includes anonymized data from Fitbit Sense wearable devices in different granularities, such as stress score, sleep duration, calories, steps, active minutes, oxygen variation, and heart rate among others.…”

“…However, the UnStressMe framework and pipeline can be generalized to other subdomains within ST (e.g., physical activity or sleep). In summary, the contributions of this work are as follows: Newly-released, in-the-wild dataset: We introduce for the first time to the community and analyze the LifeSnaps dataset [9], a recently published, multi-modal, real-world dataset of ST data, self-reported user labels and psychological and behavioral surveys, tackling limitations L1 and L2.…”