Tang, Minghu
Wang, Junjie
Bu, Xuan
Zhang, Jiayi
Luo, Peng
Funding for this research was provided by:
Special Project for the Transformation of Scientific and Technological Achievements of Qinghai Province (2025-GX-143)
Article History
Received: 26 November 2025
Accepted: 20 March 2026
First Online: 31 March 2026
Declarations
:
: Not applicable. All data used in this study are publicly available secondary datasets (e.g., Kaggle-hosted NYPD crime data, New York City TLC taxi trajectory data, Colorado School of Mines EOG nighttime light data) that do not involve human subjects, personal identifiable information, or experimental procedures requiring ethical oversight. No ethical approval or participant consent was required for the use of these public datasets.
: Not applicable. This study does not include any personal data, individual-level information, images, or content that requires explicit consent for publication.
: All authors (Tang Minghu, Wang Junjie, Bu Xuan, Zhang Jiayi, Luo Peng) declare no known competing financial interests or personal relationships that could have appeared to influence the work reported in this manuscript.
: Not applicable. The code used for model training (LLM-STT model implementation, spatiotemporal attention module, LLM lightweight fine-tuning) and data analysis (feature engineering, experimental result validation) is available from the corresponding author (Wang Junjie) upon reasonable request.
: The authors declare no competing interests.