CSER Home
» Tools: Bayesian Network Clasification
ArcGIS
Tools: Bayesian Network Classification
The Bayesian Network Classification extension links the nodes in a
Bayesian Network (BN) to attributes in a GIS. A BN is a graphical model
that represents variables (as nodes) and cause-effect relationships (as
directed links) between variables. All geographical data has uncertainty
associated with its attributes, a BN uses belief probabilities to
represent these uncertainties in a mathematically sound way. An example
of a Bayesian Network (BN) is shown below.
This is a simple example for a model of land
surface stability. The BN classification tool categorizes nodes as: background, classification and observation
variables. In the figure above, the three BN variables for slope,
soil and wetness index are influential factors
(background variables) used to derive (classification variable)
slope stability. The direct consequences of slope instability may
be observed (observation variable) from surface failures such
as soil creep or slips. The tool is designed to interpret
classification problems following this generic structure, we believe
many problems fit this form.
The Bayesian classification tool combines two powerful information
technologies. Bayesian Networks provide a complete specification for a
problem domain and are easy to visualize. GIS is a popular tool for
storing, analyzing and visualizing geographic data. However it is
difficult to store complex relationships and processes between objects
in GIS. Probabilities are well suited for representing uncertainties in
GIS data, especially environmental information and expert judgements.
Hence we see an ideal match-up between GIS and BN’s, one provides a
model of the world and other provides data about the world.
Bayesian classification tool is generic and is well suited to
applications for:
- Environmental models which combine GIS mapping and field monitoring data
- Risk assessment to assess the combination of the likely
occurrence of an environmental threat with its expected consequences
- Land suitability models which integrate information based upon
desired criteria and conditions.
The Bayesian Network Classification tool may be downloaded for free. It is installed
as an extension to ArcGIS version 9.2 (and later versions). It is developed on two software
packages: i) Netica™ from Norsys Software Corporation for working with
Bayesian belief nets is included in this install, and ii) Microsoft's
.NET v2 for the application development is included in the ArcGIS
install.
Download the manual here(568 KB) to learn more about the BN
classification tool. To obtain the software complete the form below and
submit request to download over the Internet.
ArcGIS Extension:
Bayesian Network Classification
<%
agreed=request.form("agree_to_terms")
email1=request.form("email_address")
dl_name=request.form("dl_name")
dl_company=request.form("dl_company")
if agreed="" or agreed="No" then
%>
<%
end if
if agreed="Yes" and email1>"" then
' send email to David, ifthat worked then show downloads:
set iMsg = Server.CreateObject("CDO.Message")
set iConf = Server.CreateObject("CDO.Configuration")
' Set Flds = iConf.Fields
With iConf.Fields
.Item("http://schemas.microsoft.com/cdo/configuration/sendusing") = 2
.Item("http://schemas.microsoft.com/cdo/configuration/smtpserver") = "smtp.uq.edu.au"
.Item("http://schemas.microsoft.com/cdo/configuration/smtpconnectiontimeout") = 10
.Item("http://schemas.microsoft.com/cdo/configuration/smtpserverport") = 25
.Update
End With
on error resume next
With iMsg
Set .Configuration = iConf
.To = "d.pullar@uq.edu.au"
.From = email1
.Subject = "BN Classification download"
.HTMLBody = dl_name & " downloaded BN Classification for ArcGIS " & " e-mail: " & email1 & " Company: " & dl_company
.Send
End With
if err<>0 then
response.write "Sorry, could not send the message, please check the email address! - Go Back -"
err=0
else
response.write " Thank you, " & email1 & " - Please download : "
response.write "BN Classification for ArcGIS 9.2 (and later versions)"
end if
' ----------------------------------------------------------------------
end if
%>
Contact David Pullar
d.pullar@uq.edu.au if you have
questions.
|
