Research projects

Knowledge-based AI Application in Transportation

2023 - Present

Project Overview

This pipeline adds semantic information to GPS trajectories by combining language-model-based point-of-interest classification with probabilistic activity inference. It links locations to visit purposes and supports analysis across regional datasets.

System Flow Diagram
Language-model-based point-of-interest classification and activity inference.

Key Contributions

  • Combines point-of-interest classification with probabilistic inference of travel activities.
  • Applies the framework across regional point-of-interest datasets without additional training.

Results

  • 93.4% accuracy and 96.1% F-1 score in POI classification
  • 91.7% accuracy with 92.3% F-1 score in activity inference
POI Analysis
Framework adaptability demonstration across different regional POI datasets. Example of POIs and classification results from (a) LA county and (b) Egypt.
Trajectory Analysis
Examples of semantic trajectories post-data mining: (a) Activity-annotated stay points in LA. (b) Activity timelines.

Related Publications

All research projects