ATTRIBUTES
Pratheepchanther T
I M.Sc Data Analytics
Agenda
Spatial Mining – Introduction
Spatial Databases
What is : core goal of a spatial data mining project is to distinguish the information in order to build real,
actionable patterns to present, excluding things like statistical coincidence, randomized spatial modeling or
irrelevant results. One way analysts may do this is by combing through data looking for "same-object" or "object-
equivalent" models to provide accurate comparisons of different geographic locations.
Ref: https://www.techopedia.com/definition/30595/spatial-data-mining
Spatial Data Mining Architecture
HUMAN COMPUTER INTERACTION SYSTEM
Ref:spatialdatamining-150327100401-conversion-gate01.pdf
Spatial Data - Properties
Ref:spatialdatamining-150327100401-conversion-gate01.pdf
Spatial Data - Properties - Spatial autocorrelation
Items in a traditional data are independent of each other, – whereas
properties of locations in a map are often “auto-correlated”.
Example:
• People with similar backgrounds tend to live in the same area.
• Economies of nearby regions tend to be similar
• Changes in temperature occur gradually over space
Ref:spatialdatamining-150327100401-conversion-gate01.pdf
Spatial Data - Properties - Spatial heterogeneity
• Auto correlation
• Patterns usually have to be defined in the spatial attribute subspace
and not in the complete attribute space
• Longitude and latitude (or other coordinate systems) are the glue that
link different data collections together.
• People are used to maps in GIS therefore, data mining results have to
summarized on the top of maps.
• Patterns not only refer to points, but can also refer to lines, or
polygons or other higher order geometrical objects
Ref:spatialdatamining-150327100401-conversion-gate01.pdf
Spatial Data - Properties - Spatial Relations
Spatial databases do not store spatial relations explicitly – Additional functionality
required to compute them
Three types of spatial relations specified by the Open Geospatial Consortium (OGC)
reference model
– Distance relations
Euclidean distance between two spatial features
Ref:spatialdatamining-150327100401-conversion-gate01.pdf
Spatial Databases
Stores a large amount of space-related data Few Spatial databases:
• Maps SQL, MySQL(geometry),
• Remote Sensing Neo4J, IBM Informix,
Oracle Spatial, H2, SAP
• Medical Imaging HANA,GeoMesa
• VLSI chip layout (Apache’s spatio-
temporal database)
Ref:1) spatialdatamining-150327100401-conversion-gate01.pdf
2) https://en.wikipedia.org/wiki/Spatial_database
Spatial Mining - Attributes
Spatial Information - are generally multi-dimensional and correlated
• Spatial attribute: geographically referenced
• Neighbourhood and extent
• Location, e.g., longitude, latitude, elevation
Ref:spatialdatamining-150327100401-conversion-gate01.pdf
Spatial Mining - Attributes
Discrete attributes
Has only a finite or countable infinite set of values
Example:
• Zip codes,
• Count or set of words in a collection of documents
• Often represented as integer variables
• Note: Binary attributes are a special case of discrete attributes
Continuous attributes
Has real numbers as attribute values. Real values can only be measured and
represented using finite number of digits and typically represented as floating
point variables
Example:
• Temperature
• Height or Weight
Ref:http://user.it.uu.se/~kostis/Teaching/DM-05/Slides/lecture02.pdf
Spatial Mining - Attribute values
Numbers or Symbols assigned to an attribute
Attribute v/s Attribute values
Same attribute can be mapped to different values
Example: Height can be measured in feet or meters
Ref: http://www.jaist.ac.jp/~bao/talks/SpatialDM.pdf
Spatial Mining - Attribute types
Ref: https://sqlmentalist.com/2012/06/28/bisql-100-sql-server-with-spatial-dataopportunity-use-new-features-and-example-with-few-queries/
SPATIAL ATTRIBUTE DATA
Ref: https://pediaa.com/what-is-the-difference-between-attribute-data-and-spatial-data/
Spatial Mining - Data types
Ref: https://sqlmentalist.com/2012/06/28/bisql-100-sql-server-with-spatial-dataopportunity-use-new-features-and-example-with-few-queries/
Spatial Mining - Attribute types
Attribute
Description Transformation Operation Comments Example
type
The values of a nominal attribute
Nominal
are just different names, i.e., mode, entropy, If all employee ID numbers zip codes, employee ID
Any permutation of
nominal attributes provide only contingency were reassigned, would it make numbers, eye color,
values
enough information to distinguish correlation, χ2 test any difference? sex: {male, female}
one object from another. (=, ≠)
An order preserving median, An attribute encompassing the
hardness of minerals,
Ordinal
The values of an ordinal attribute change of values, i.e., percentiles, rank notion of good, better
{good, better, best},
provide enough information to new_value = correlation, run best can be represented
grades, street numbers
order objects. (<, >) f(old_value) where f is tests, sign tests equally well by the values {1, 2,
a monotonic function. 3} or by {0.5, 1, 10}.
mean, standard
Thus the Fahrenheit and
For interval attributes, the new_value =a * deviation,
Interval
temperature in Kelvin,
geometric mean,
For ratio variables, both monetary quantities,
Ratio
Ref: http://www.jaist.ac.jp/~bao/talks/SpatialDM.pdf
Thank you