Geography Reference
In-Depth Information
17.2
Data Describing ST Phenomena
In this section we identify the various data types for describing ST change
phenomena that can be found in literature and statistical data models. Table
17.1
shows a summary of input data types for change pattern mining.
17.2.1
Temporal Data
Figure
17.2
a shows a time series from climate science. It is a plot of the Sahel
rainfall index (JISAO
2011
), a normalized rainfall from June to October over the
Sahel region in Africa, over the past 111 years. Change pattern analysis on such
datasets is of interest to climate change researchers.
17.2.1.1
Event Reports
Event reports are collections of time-stamped data instances. Unlike time series,
event report data may not be uniformly distributed in time, and usually have spatial
dimensions such as the locations of the events. Disease reports in public health
and crime reports in public safety are typical event report data. (Event reports are
discussed further in Sect.
17.2.3
)
17.2.2
Spatial Data
17.2.2.1
Point Data
Point data on two dimensional space usually represents the locations of a collection
of events. Each point is associated with a pair of coordinates. In a public health
Table 17.1
Input data types for change pattern mining and corresponding research area
Data type
Spatial statistics
Time series analysis
Image processing
Temporal
Time series
x
Spatial
Point data
x
Polygon data
x
Lines and paths
x
Spatial field
x
Spatio-temporal
ST event reports
x
Image sequences
x
ST raster data
x
x
Search WWH ::
Custom Search