EUVIMEDEUVIMED entdecken
Crossref

CLIFT: Cross-City Mobility-Derived Lifestyle Pattern Transfer for Improved Multi-City Next Location Prediction

Haru Terashima, Naoki Tamura, Kazuyuki Shoji, Tahera Hossain, Shin Katayama, Kenta Urano, Takuro Yonezawa, Nobuo Kawaguchi

ACM Transactions on Spatial Algorithms and Systems · 30.06.2026 · 12 (2) · 1-24

DOI: 10.1145/3811037 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Spatial Algorithms and Systems, online erschienen am 06.05.2026. Verfasst von Haru Terashima und 7 weiteren Autor:innen. Die Arbeit nennt 49 Literaturangaben. Gefördert von: JST CREST, JST RISTEX, NEDO SIP3.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We propose CLIFT ( C ross-City Lif estyle Pattern T ransfer for Human Mobility Prediction), a novel framework that enhances human mobility prediction by integrating general lifestyle patterns shared across cities with city-specific mobility patterns. Accurate human mobility prediction in urban environments is critical for transportation planning, marketing strategies, and disaster response. However, most existing deep learning approaches use only single-city data and exhibit significant performance degradation in small cities with limited training data. These limitations motivate methods that jointly leverage cross-city behavioral patterns and city-specific mobility characteristics. CLIFT addresses this challenge through dual complementary encoders: one captures general lifestyle patterns shared across cities, and the other captures city-specific mobility patterns; their outputs are integrated with a Transformer-based mobility predictor (LP-BERT). This architecture enables the model to jointly capture cross-city transferable behavioral patterns and city-specific mobility characteristics. We evaluated the effectiveness of CLIFT through experiments on the multi-city human mobility dataset LYMob-4Cities, comparing its performance with both single-city and multi-city deep learning-based methods. On average, CLIFT improved GEOBLEU and Top-1 accuracy by 11.1% and 10.6% over the single-city baseline, and by 5.0% and 7.9% over the multi-city baseline, respectively. Furthermore, CLIFT outperformed the top-ranked teams in the international competition, Human Mobility Prediction Challenge 2024 , demonstrating superior predictive performance under the same dataset and task setting.

Förderung

  • JST CREST
  • JST RISTEX
  • NEDO SIP3
  • JSPS KAKENHI

Zitieren

APA

Terashima, H., Tamura, N., Shoji, K., Hossain, T., Katayama, S., Urano, K., Yonezawa, T., & Kawaguchi, N. (2026). CLIFT: Cross-City Mobility-Derived Lifestyle Pattern Transfer for Improved Multi-City Next Location Prediction. ACM Transactions on Spatial Algorithms and Systems, 12(2), 1-24. https://doi.org/10.1145/3811037

Vancouver

Terashima H, Tamura N, Shoji K, Hossain T, Katayama S, Urano K, et al. CLIFT: Cross-City Mobility-Derived Lifestyle Pattern Transfer for Improved Multi-City Next Location Prediction. ACM Transactions on Spatial Algorithms and Systems. 2026;12(2):1-24. doi:10.1145/3811037

BibTeX
@article{Terashima2026,
  author = {Terashima, Haru and Tamura, Naoki and Shoji, Kazuyuki and Hossain, Tahera and Katayama, Shin and Urano, Kenta and Yonezawa, Takuro and Kawaguchi, Nobuo},
  title = {CLIFT: Cross-City Mobility-Derived Lifestyle Pattern Transfer for Improved Multi-City Next Location Prediction},
  journal = {ACM Transactions on Spatial Algorithms and Systems},
  year = {2026},
  volume = {12},
  number = {2},
  pages = {1--24},
  doi = {10.1145/3811037}
}
Kostenlos starten →