Home
This Title All WIREs
WIREs RSS Feed
How to cite this WIREs title:
WIREs Data Mining Knowl Discov
Impact Factor: 2.111

Spatiotemporal change footprint pattern discovery: an inter‐disciplinary survey

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

Given a definition of change and a dataset about spatiotemporal (ST) phenomena, ST change footprint discovery is the process of identifying the location and/or time of such changes from the dataset. Change footprint discovery is fundamentally important for the study of climate change, the tracking of disease, and many other applications. Methods for detecting change footprints have emerged from a diverse set of research areas, ranging from time series analysis and remote sensing to spatial statistics. Researchers have much to learn from one another, but are stymied by inconsistent use of terminology and varied definitions of change across disciplines. Existing reviews focus on discovery methods for only one or a few types of change footprints (e.g., point change in a time series). To facilitate sharing of insights across disciplines, we conducted a multi‐disciplinary review of ST change patterns and their respective discovery methods. We developed a taxonomy of possible ST change footprints and classified our review findings accordingly. This exercise allowed us to identify gaps in the research that we consider ripe for exploration, most notably change pattern discovery in vector ST datasets. In addition, we illustrate how such pattern discovery might proceed using two case studies from historical GIS. WIREs Data Mining Knowl Discov 2014, 4:1–23. doi: 10.1002/widm.1113

Conflict of interest: The authors have declared no conflicts of interest for this article.

A flow chart showing ST change footprint pattern discovery process.
[ Normal View | Magnified View ]
U.S. railroad network expansion in the 19th century (best viewed in color). (a) U.S. railroad network in 1840. (b) U.S. railroad network in 1850. (c) U.S. railroad network in 1861. (d) U.S. railroad network in 1870.
[ Normal View | Magnified View ]
A comparison of county seats locations at 300 and 400 AD in China (best viewed in color). (a) County seats of China in 300 AD. (b) County seats of China in 400 AD. (c) Change of county seat locations between 300 AD and 400 AD.
[ Normal View | Magnified View ]
An example of network footprint to summarize high activity patterns. (a) A spatial network with activities. (b) Polygon footprint to summarize activities with Euclidian distance (by CrimeStat) (c) Polygon footprint to summarize activities with network distance (by CrimeStat). (d) Network footprint to summarize activities.
[ Normal View | Magnified View ]
Input and output example of spatial scan statistics. (a) A point process dataset. (b) The mostly likely cluster discovered
[ Normal View | Magnified View ]
Change boundary (line) footprints on the world GDP growth data.
[ Normal View | Magnified View ]
An example of pixel‐wise change detection outputs. (a) Impervious surface image of an area in 1986. (b) Impervious surface image of the same area in 1991. (c) Locations with difference exceeding 60% of the maximum magnitude between (a) and (b).
[ Normal View | Magnified View ]
Example of interval change on a time series found by the interesting sub‐path discovery method.
[ Normal View | Magnified View ]
Example of time point change identified by time series segmentation.
[ Normal View | Magnified View ]
A time point change footprint pattern discovered by CUSUM.
[ Normal View | Magnified View ]
An example of spatial zonal change footprint (best viewed in color). (a) Vegetation cover (in NDVI) in Africa, August, 1981. (b) Footprints of spatial zonal change patterns with longitudinal changes in vegetation cover of Africa.
[ Normal View | Magnified View ]
An example of lattice Wombling results on a raster field (best viewed in color).
[ Normal View | Magnified View ]
Three sets of results for one time series dataset using three different definitions of change. (a) Statistical parameter change in a time series. (b) Value change in a time series. (c) Change in model fitted on a time series.
[ Normal View | Magnified View ]
A spatial vector dataset showing GDP growth in countries around the world in 2011 (best viewed in color).
[ Normal View | Magnified View ]
A spatial raster dataset showing vegetation cover (in NDVI value) of Africa (best viewed in color).
[ Normal View | Magnified View ]
An example of time series from climate science.
[ Normal View | Magnified View ]

Browse by Topic

Fundamental Concepts of Data and Knowledge > Motivation and Emergence of Data Mining
Technologies > Structure Discovery and Clustering
Algorithmic Development > Spatial and Temporal Data Mining

Access to this WIREs title is by subscription only.

Recommend to Your
Librarian Now!

The latest WIREs articles in your inbox

Sign Up for Article Alerts