Modeling Activity Tracker Data Using Deep Boltzmann Machines

TL;DR

This paper explores the use of deep Boltzmann machines to model unlabeled activity tracker data, revealing distinct weekly usage patterns and demonstrating the method's feasibility for health-related data analysis.

Contribution

It introduces a novel application of deep Boltzmann machines for modeling activity tracker data and shows their effectiveness in uncovering usage patterns.

Findings

Two distinct weekly usage patterns identified

DBMs can generate artificial samples of activity data

Feasibility of deep learning for health data modeling

Abstract

Commercial activity trackers are set to become an essential tool in health research, due to increasing availability in the general population. The corresponding vast amounts of mostly unlabeled data pose a challenge to statistical modeling approaches. To investigate the feasibility of deep learning approaches for unsupervised learning with such data, we examine weekly usage patterns of Fitbit activity trackers with deep Boltzmann machines (DBMs). This method is particularly suitable for modeling complex joint distributions via latent variables. We also chose this specific procedure because it is a generative approach, i.e., artificial samples can be generated to explore the learned structure. We describe how the data can be preprocessed to be compatible with binary DBMs. The results reveal two distinct usage patterns in which one group frequently uses trackers on Mondays and Tuesdays, whereas the other uses trackers during the entire week. This exemplary result shows that DBMs are feasible and can be useful for modeling activity tracker data.