Big Trajectory Data Mining: A Survey of Methods, Applications, and Services

Area of Science:

• Urban computing
• Data science
• Geospatial analysis

Background:

• Smart infrastructure and location-aware terminals generate rich spatiotemporal trajectory data.
• Trajectory data analysis aids in understanding urban dynamics for problem-solving in transportation, environment, and public security.
• Existing research is fragmented, lacking a comprehensive overview of methods, applications, and generated services.

Purpose of the Study:

• To provide an integrated review of trajectory data mining methods and applications.
• To classify trajectory data applications into social, traffic, and operational dynamics.
• To guide researchers and practitioners in selecting appropriate methods for urban challenges.

Main Methods:

• Discusses relationships among prevailing trajectory mining methods.
• Classifies trajectory data applications into three major groups: social dynamics, traffic dynamics, and operational dynamics.
• Reviews existing studies to identify tackled problems, tested methods, and generated services.

Main Results:

• Identifies key trajectory mining methods and their interrelationships.
• Categorizes applications into social, traffic, and operational dynamics, highlighting urban reality insights.
• Discusses potential services derivable from trajectory data analysis.

Conclusions:

• Highlights the need for an integrated approach to trajectory data mining.
• Offers practical implications for data analysts and planners in method selection.
• Aims to inspire further research by identifying gaps in current methods and applications.