Using Topological Analysis to Support Event-Guided Exploration in Urban Data.

Area of Science:

• Urban Informatics
• Computational Topology
• Data Science

Background:

• Urban environments generate vast amounts of complex, spatio-temporal data.
• Manual exploration of this data is often inefficient and impractical for policy and administration.
• Understanding urban dynamics requires analyzing data across multiple scales and time periods.

Purpose of the Study:

• To develop a novel technique for event-guided exploration of large-scale urban datasets.
• To enable efficient analysis and understanding of complex urban dynamics.
• To support data-driven decision-making for urban policy and public services.

Main Methods:

• Modeling urban data as time-varying scalar functions.
• Utilizing computational topology to automatically identify significant events within data slices.
• Developing algorithms for grouping, indexing, and interactively querying identified events.
• Creating a visual exploration interface to guide users toward relevant data.

Main Results:

• Demonstrated effectiveness of the event-guided exploration technique on New York City (NYC) taxi and subway data.
• Successfully identified and presented significant events and trends in urban datasets.
• Facilitated interactive exploration and querying of spatio-temporal urban data patterns.

Conclusions:

• The proposed event-guided exploration technique significantly enhances the analysis of complex urban data.
• This approach offers a practical solution for urban planners and administrators to leverage big data.
• Improved understanding of urban dynamics can lead to better public services and reduced environmental impact.