DySTAN: Joint Modeling of Sedentary Activity and Social Context from Smartphone Sensors

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Sneh, Aditya, Sahu, Nilesh Kumar, Gupta, Snehil, Lone, Haroon R.
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914176316735488
author Sneh, Aditya
Sahu, Nilesh Kumar
Gupta, Snehil
Lone, Haroon R.
author_facet Sneh, Aditya
Sahu, Nilesh Kumar
Gupta, Snehil
Lone, Haroon R.
contents Accurately recognizing human context from smartphone sensor data remains a significant challenge, especially in sedentary settings where activities such as studying, attending lectures, relaxing, and eating exhibit highly similar inertial patterns. Furthermore, social context plays a critical role in understanding user behavior, yet is often overlooked in mobile sensing research. To address these gaps, we introduce LogMe, a mobile sensing application that passively collects smartphone sensor data (accelerometer, gyroscope, magnetometer, and rotation vector) and prompts users for hourly self-reports capturing both sedentary activity and social context. Using this dual-label dataset, we propose DySTAN (Dynamic Cross-Stitch with Task Attention Network), a multi-task learning framework that jointly classifies both context dimensions from shared sensor inputs. It integrates task-specific layers with cross-task attention to model subtle distinctions effectively. DySTAN improves sedentary activity macro F1 scores by 21.8% over a single-task CNN-BiLSTM-GRU (CBG) model and by 8.2% over the strongest multi-task baseline, Sluice Network (SN). These results demonstrate the importance of modeling multiple, co-occurring context dimensions to improve the accuracy and robustness of mobile context recognition.
format Preprint
id arxiv_https___arxiv_org_abs_2512_02025
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DySTAN: Joint Modeling of Sedentary Activity and Social Context from Smartphone Sensors
Sneh, Aditya
Sahu, Nilesh Kumar
Gupta, Snehil
Lone, Haroon R.
Signal Processing
Artificial Intelligence
Human-Computer Interaction
Machine Learning
Accurately recognizing human context from smartphone sensor data remains a significant challenge, especially in sedentary settings where activities such as studying, attending lectures, relaxing, and eating exhibit highly similar inertial patterns. Furthermore, social context plays a critical role in understanding user behavior, yet is often overlooked in mobile sensing research. To address these gaps, we introduce LogMe, a mobile sensing application that passively collects smartphone sensor data (accelerometer, gyroscope, magnetometer, and rotation vector) and prompts users for hourly self-reports capturing both sedentary activity and social context. Using this dual-label dataset, we propose DySTAN (Dynamic Cross-Stitch with Task Attention Network), a multi-task learning framework that jointly classifies both context dimensions from shared sensor inputs. It integrates task-specific layers with cross-task attention to model subtle distinctions effectively. DySTAN improves sedentary activity macro F1 scores by 21.8% over a single-task CNN-BiLSTM-GRU (CBG) model and by 8.2% over the strongest multi-task baseline, Sluice Network (SN). These results demonstrate the importance of modeling multiple, co-occurring context dimensions to improve the accuracy and robustness of mobile context recognition.
title DySTAN: Joint Modeling of Sedentary Activity and Social Context from Smartphone Sensors
topic Signal Processing
Artificial Intelligence
Human-Computer Interaction
Machine Learning
url https://arxiv.org/abs/2512.02025