Modeling Interdependent and Periodic Real-World Action Sequences

TL;DR

This paper introduces a novel statistical model for predicting human actions in health-related applications by capturing interdependencies, periodicities, and time-varying behaviors, significantly improving prediction accuracy over existing methods.

Contribution

The work presents a personalized multivariate temporal point process model that jointly captures action dependencies and periodicities, advancing beyond prior models focused mainly on item consumption.

Findings

Improves action prediction accuracy by up to 156% over existing methods.

Effectively models rare and periodic actions like walking and biking.

Demonstrates applicability on large-scale real-world activity datasets.

Abstract

Mobile health applications, including those that track activities such as exercise, sleep, and diet, are becoming widely used. Accurately predicting human actions is essential for targeted recommendations that could improve our health and for personalization of these applications. However, making such predictions is extremely difficult due to the complexities of human behavior, which consists of a large number of potential actions that vary over time, depend on each other, and are periodic. Previous work has not jointly modeled these dynamics and has largely focused on item consumption patterns instead of broader types of behaviors such as eating, commuting or exercising. In this work, we develop a novel statistical model for Time-varying, Interdependent, and Periodic Action Sequences. Our approach is based on personalized, multivariate temporal point processes that model time-varying action propensities through a mixture of Gaussian intensities. Our model captures short-term and long-term periodic interdependencies between actions through Hawkes process-based self-excitations. We evaluate our approach on two activity logging datasets comprising 12 million actions taken by 20 thousand users over 17 months. We demonstrate that our approach allows us to make successful predictions of future user actions and their timing. Specifically, our model improves predictions of actions, and their timing, over existing methods across multiple datasets by up to 156%, and up to 37%, respectively. Performance improvements are particularly large for relatively rare and periodic actions such as walking and biking, improving over baselines by up to 256%. This demonstrates that explicit modeling of dependencies and periodicities in real-world behavior enables successful predictions of future actions, with implications for modeling human behavior, app personalization, and targeting of health interventions.