1 / 14

Zechun Cao

A Methodology for Finding Uniform Regions in Spatial Data and its Application to Analyzing the Composition of Cities. Zechun Cao. UH DMML Advisor: Dr. Eick. Talk Organization. Introduction Methodology Preliminary Results Future Work. Urban Computing.

dot
Download Presentation

Zechun Cao

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. A Methodology for Finding Uniform Regions in Spatial Data and its Application to Analyzing the Composition of Cities Zechun Cao UH DMML Advisor: Dr. Eick

  2. Talk Organization Introduction Methodology Preliminary Results Future Work UH-DMML

  3. Urban Computing • Serves for the modern cities with rapid progress of urbanization and civilization. • Aims to understand the nature and science behind the phenomenon. UH-DMML

  4. Research Tasks • Analyzes the composition of the buildings in a city. • Develops framework that captures the spatial heterogeneity by non-spatial attributes. UH-DMML

  5. Research Goals • Partitions a given space into uniform regions based on a domain expert’s notion of uniformity by maximizing a plug-in measure of uniformity. • Provides analysis functions to create summaries for the identified uniform regions. UH-DMML

  6. CLEVER UH-DMML

  7. Interestingness • Purity • Low Variance UH-DMML

  8. Interestingness • Building Type Signature (90%, 5%, 1%, 3%, 0%, 1% ) UH-DMML

  9. Dataset UH-DMML

  10. Preliminary Results Building Type Purity Experiment Cluster#0: Commercial Area UH-DMML

  11. Preliminary Results Building Size Distribution Experiment • 30 out of 36 clusters has the area standard deviation lower than the dataset standard deviation. UH-DMML

  12. Preliminary Results Query Spatial Dataset by Signature UH-DMML

  13. Future Work • Sensitivity analysis of initialization. • Merging the clusters across multiple runs to gain better clustering quality. • Parallel spatial data mining algorithms. UH-DMML

  14. Thank you!

More Related