Stay-Contextual Trigger: A Confidence-Based Algorithm for Energy-Efficient and Behaviorally Grounded Mobile Notifications
Complex Systems Informatics and Modeling Quarterly · 31.10.2026 · (48) · 33-51
DOI: 10.7250/csimq.2026-48.02 ↗Kurz gefasst
Fachartikel in der Zeitschrift Complex Systems Informatics and Modeling Quarterly, online erschienen am 31.10.2026. Verfasst von Yudhy Setyo Purwanto und 2 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Context-aware mobile applications often trigger notifications based solely on geofence entry, leading to high battery consumption and disruptive alerts during transient movement. To address this, we propose the Stay-Contextual Trigger (SCT) framework, which gates high-power GNSS activation behind low-power activity recognition and enforces a context-adaptive dwell-time requirement before delivering any notification. We evaluate SCT through a 14-day field study with 60 compliant Android users (Per-Protocol analysis after attrition filtering), comparing it against standard geofencing in four real-world outdoor zones. The SCT system computes a Contextual Readiness Score via a Confidence-Based Weighted Algorithm (CBWA), dynamically adjusting the weights of activity, proximity, dwell time, and behavioral history based on location semantics. Results show that SCT reduces daily battery drain by 40.3% (from 19.30% to 11.52%), increases notification acceptance from 32.5% to 89.4%, and achieves a System Usability Scale score of 81.25, an 18.8-point improvement over the control group’s 62.42. The high acceptance rate (89.4%) further validates sustained user engagement, ensuring metrics reflect consistent usage. These findings demonstrate that behavioral readiness, not mere proximity, is essential for sustainable and user-respectful mobile systems.
Zitieren
APA
Purwanto, Y. S., Gernowo, R., & Nugraheni, D. M. K. (2026). Stay-Contextual Trigger: A Confidence-Based Algorithm for Energy-Efficient and Behaviorally Grounded Mobile Notifications. Complex Systems Informatics and Modeling Quarterly, 33-51. https://doi.org/10.7250/csimq.2026-48.02
Vancouver
Purwanto YS, Gernowo R, Nugraheni DMK. Stay-Contextual Trigger: A Confidence-Based Algorithm for Energy-Efficient and Behaviorally Grounded Mobile Notifications. Complex Systems Informatics and Modeling Quarterly. 2026:33-51. doi:10.7250/csimq.2026-48.02
BibTeX
@article{Purwanto2026,
author = {Purwanto, Yudhy Setyo and Gernowo, Rahmat and Nugraheni, Dinar Mutiara Kusumo},
title = {Stay-Contextual Trigger: A Confidence-Based Algorithm for Energy-Efficient and Behaviorally Grounded Mobile Notifications},
journal = {Complex Systems Informatics and Modeling Quarterly},
year = {2026},
number = {48},
pages = {33--51},
doi = {10.7250/csimq.2026-48.02}
}
EUVIMED