Statistical counterpoint: Knowledge discovery of choreographic information using spatio-temporal analysis and visualization

This work presents work to develop a rich data base on choreographic information and use it as a test-bed for existing geovisual approaches to temporal data, demonstrating the ability of linked, multivariate displays to pull out distinct differences and similarities in activity patterns and temporal clustering.

Abstract

Abstract Temporal GIS has rapidly moved from concepts and methods that took its inspiration based on theories largely developed in a data-poor setting, to a situation where tools and theories are needed for large, individual-level, spatio-temporally detailed data sets. We present work to develop a rich data base on choreographic information and use it as a test-bed for existing geovisual approaches to temporal data. Our visual analysis demonstrates the ability of linked, multivariate displays to pull out distinct differences and similarities in activity patterns and temporal clustering. We also include illustrative animations that identify some practical limitations of existing exploratory software in dealing with large volumes of spatio-temporal data.