TGT: A Temporal Gating Transformer for Smartphone App Usage Prediction

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Li, Longlong, Qu, Cunquan, Wang, Guanghui
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909811432488960
author Li, Longlong
Qu, Cunquan
Wang, Guanghui
author_facet Li, Longlong
Qu, Cunquan
Wang, Guanghui
contents Accurately predicting smartphone app usage is challenging due to the sparsity and irregularity of user behavior, especially under cold-start and low-activity conditions. Existing approaches mostly rely on static or attention-only architectures, which struggle to model fine-grained temporal dynamics. We propose TGT, a Transformer framework equipped with a temporal gating module that conditions hidden representations on the hour-of-day. Unlike conventional time embeddings, temporal gating adaptively rescales feature dimensions in a time-aware manner, working orthogonally to self-attention and strengthening temporal sensitivity. TGT further incorporates a context-aware encoder that integrates session sequences and user profiles into a unified representation. Experiments on two real-world datasets, Tsinghua App Usage and LSApp, demonstrate that TGT significantly outperforms 15 competitive baselines, achieving notable gains in HR@1 and maintaining robustness under cold-start scenarios. Beyond accuracy, analysis of gating vectors uncovers interpretable daily usage rhythms, showing that TGT learns human-consistent patterns of app behavior. These results establish TGT as both a powerful and interpretable framework for time-aware app usage prediction.
format Preprint
id arxiv_https___arxiv_org_abs_2502_16957
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TGT: A Temporal Gating Transformer for Smartphone App Usage Prediction
Li, Longlong
Qu, Cunquan
Wang, Guanghui
Machine Learning
Accurately predicting smartphone app usage is challenging due to the sparsity and irregularity of user behavior, especially under cold-start and low-activity conditions. Existing approaches mostly rely on static or attention-only architectures, which struggle to model fine-grained temporal dynamics. We propose TGT, a Transformer framework equipped with a temporal gating module that conditions hidden representations on the hour-of-day. Unlike conventional time embeddings, temporal gating adaptively rescales feature dimensions in a time-aware manner, working orthogonally to self-attention and strengthening temporal sensitivity. TGT further incorporates a context-aware encoder that integrates session sequences and user profiles into a unified representation. Experiments on two real-world datasets, Tsinghua App Usage and LSApp, demonstrate that TGT significantly outperforms 15 competitive baselines, achieving notable gains in HR@1 and maintaining robustness under cold-start scenarios. Beyond accuracy, analysis of gating vectors uncovers interpretable daily usage rhythms, showing that TGT learns human-consistent patterns of app behavior. These results establish TGT as both a powerful and interpretable framework for time-aware app usage prediction.
title TGT: A Temporal Gating Transformer for Smartphone App Usage Prediction
topic Machine Learning
url https://arxiv.org/abs/2502.16957