Exploring Spatiotemporal Neighborhood Dynamics and Relationships with Time Series Clustering

Authors: Carlos Baez*, University of California, Santa Barbara
Topics: Spatial Analysis & Modeling, Temporal GIS, Urban Geography
Keywords: urban data science, spatiotemporal analysis, transportation
Session Type: Paper
Day: 4/6/2020
Start / End Time: 1:30 PM / 2:45 PM
Room: Century, Sheraton, IM Pei Tower, Mezzanine Level
Presentation File: No File Uploaded

Time series clustering is a useful exploratory tool for investigating the similarities among a set of time
series. In this presentation we consider taxi trip count time series of New York City and how time series
clustering can be used to describe and explore the relationship between different neighborhoods. We
first present different model, feature, and shape-based approaches along with various similarity
measures corresponding to each approach. Then we show how spatial and/or temporal aggregation
affects these types of analyses and discuss the implications with respect to the modifiable
spatial/temporal unit problem.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login