A review of quantitative methods for movement data

Abstract

The collection, visualization, and analysis of movement data is at the forefront of geographic information science research. Movement data are generally collected by recording an object's spatial location (e.g., XY coordinates) at discrete time intervals. Methods for extracting useful information, for example space-time patterns, from these increasingly large and detailed datasets have lagged behind the technology for generating them. In this article we review existing quantitative methods for analyzing movement data. The objective of this article is to provide a synthesis of the existing literature on quantitative analysis of movement data while identifying those techniques that have merit with novel datasets. Seven classes of methods are identified: 1) time geography, 2) path descriptors, 3) similarity indices, 4) pattern and cluster methods, 5) individual-group dynamics, 6) spatial field methods, and 7) spatial range methods. Challenges routinely faced in quantitative analysis of movement data include difficulties with handling space and time attributes together, representing time in GIS, and using classic statistical testing procedures with space-time movement data. Areas for future research include: investigating equivalent distance comparisons in space and time, measuring interactions between moving objects, development of predictive frameworks for movement data, integrating movement data with existing geographic layers, and incorporating theory from time geography into movement models. In conclusion, quantitative analysis of movement data is an active research area with tremendous opportunity for new developments and methods.

Spatiotemporal enabled Content-based Image Retrieval

“…Time-based methods conceptualize a space-time path as multidimensional time series data. It includes synchronized Euclidean distance, Fréchet distance, dynamic time warping (DTW) and longest common sub-sequences (LCSS) (See Long and Nelson (2013) for the review of these measures). The applications of path similarity in transportation studies include clustering methods for movement trajectories, comparison of individual mobility patterns, and querying for paths in a database that are similar to a reference path.…”

Generating Heat Maps of Popular Routes Online from Massive Mobile Sports Tracking Application Data in Milliseconds While Respecting Privacy

“…One specific method uses clustering techniques; for a review of different clustering methods, see [17]. This can be done, for example, using kernel methods [18], where the visualization is based on the total number of tracks at a given location.…”

Modeling and Querying Moving Objects with Social Relationships

“…The user has some online relationships in social networks, such as Facebook, Twitter or Foursquare. In so called location based social networks (LBSN) [16][17][18], there have been a bulk of query requests concerning not only movements with spatial and temporal characteristics, but also dynamic variation of social relationships:…”