Published October 23, 2025 | Version v11
Dataset Open

Predicting Human Mobility Using Dense Smartphone GPS Trajectories and Transformer Models

  • 1. ROR icon Duke University
  • 2. ROR icon University of California, San Diego
  • 3. ROR icon Duke Kunshan University

Description

This repository contains the code and resources used in the paper titled 'Detecting Transportation Mode Using Dense Smartphone GPS Trajectories and Transformer Models.' It includes data preprocessing scripts, custom Transformer and LSTM architectures, as well as training and evaluation pipelines.

Files

geolife_processed.csv

Files (31.6 GB)

Name Size Download all
md5:811780a8adce7fd9b472c8d22aa5e073
273.7 MB Preview Download
md5:e6fd837fb1ea622f7fc7abbb798f4bcb
9.5 MB Preview Download
md5:089c11627b9cc479cbf4fc910e2571ec
31.3 GB Preview Download
md5:98301c08619d01123c044bc6d882d80a
25.8 MB Preview Download