A Review of GPS Trajectories Classification Based on Transportation Mode.

Area of Science:

• * Geographic Information Science
• * Transportation Science
• * Urban Studies

Background:

• * Global Positioning System (GPS) trajectories offer valuable insights into human activities and movement patterns.
• * Research in location-based services, transportation, and urban studies increasingly utilizes GPS data.
• * Acquisition and processing of GPS data are critical for extracting meaningful information.

Purpose of the Study:

• * To systematically review and classify existing methods for GPS trajectory data based on transportation mode.
• * To differentiate classification strategies for single-mode and mixed-mode GPS data.
• * To identify challenges and future research directions in GPS trajectory classification.

Main Methods:

• * Macroscopic classification of GPS data into single-mode and mixed-mode based on transportation type.
• * Review of existing classification methods for both single-mode and mixed-mode GPS trajectories.
• * Analysis of the contributions and unresolved issues of current classification strategies.

Main Results:

• * GPS trajectories can be broadly classified into single-mode (one transportation type) and mixed-mode (multiple transportation types).
• * Classification strategies differ significantly based on whether transition points or activity episodes are identified first.
• * Existing methods for both single-mode and mixed-mode classification have specific contributions and limitations.

Conclusions:

• * A clear understanding of single-mode versus mixed-mode GPS data classification is essential for accurate transportation mode detection.
• * The review provides a foundation for researchers to select and improve appropriate GPS trajectory classification methods.
• * Future research should address the identified unresolved issues to enhance the accuracy and applicability of transportation mode classification.