Data Mining Processes - Villanova University
Download
Report
Transcript Data Mining Processes - Villanova University
Using JMP for the Case
Competition
1-2
Overview of Case Analysis
• If you have not had formal coursework in data
mining, in order to compete in the case, you will
probably want to do the following:
• Install JMP
• Learn the basics of JMP
• Learn about partitioning the data set
(training, validation, test sets)
• Learn about specifying the type of variables
(nominal, ordinal, categorical)
1-3
Overview of Case Analysis
• Learn about specific modeling techniques
like:
• Logistic Regression
• Decision Trees
• Bootstrap Forest
• Boosted Trees
• Neural Net Models
1-4
Installing JMP
Villanova owns a site license for JMP so that
every student can install JMP at:
https://software.villanova.edu/
Enter you Villanova user id and password (keep
the organization box blank). Windows users will
select JMP 11 and Mac users will select JMP 11
(OSX). The functionality is the same in both
versions but there are some differences in
navigation and menuing.
1-5
JMP Tutorials
Two suggested tutorials for students (JMP
for Students 1 and JMP for Students 2) can
be found at:
http://www.jmp.com/about/events/ondemand
/ondemandseries.shtml?series=academic&sortOrder=d
escending
1-6
On-Demand Webcasts
• There are many On-Demand Webcasts on the
JMP website that can be viewed at:
• http://www.jmp.com/about/events/ondemand/
• Some that may be of interest to you are in
Mastering JMP:
• http://www.jmp.com/about/events/ondemand/
ondemandseries.shtml?series=mastering&sortOrder=de
scending
1-7
On-Demand Webcasts
• Additional webcasts are in Building Better
Models:
• http://www.jmp.com/about/events/ondemand/
ondemandseries.shtml?series=buildBetterModels&sort
Order=ascending
1-8
On-Demand Webcasts
• Here is a good place to get started. There are
several good webcasts that provide a good
overview of data mining, provide a discussion of
data partitioning, explain where to access the
sample data sets within JMP, and provide an
introduction to building predictive models.
• These videos can be found at:
• http://www.jmp.com/about/events/ondemand/
ondemandviewer.shtml?reglink=70130000001rBqo&seri
es=mastering
1-9
On-Demand Webcasts
• Examples of regression videos can be found at:
• http://www.jmp.com/about/events/ondemand/
ondemandviewer.shtml?reglink=701a0000000tKKg&ser
ies=buildBetterModels
• Examples of decision tree videos can be found
at:
• http://www.jmp.com/about/events/ondemand/
ondemandviewer.shtml?reglink=701a0000000tKKl&seri
es=buildBetterModels
1-10
On-Demand Webcasts
• Bootstrap Forest and Boosted Tree videos can
be found at:
• http://www.jmp.com/about/events/ondemand/
ondemandviewer.shtml?reglink=701a0000000tKKq&ser
ies=buildBetterModels
• Neural Net videos can be found at:
• http://www.jmp.com/about/events/ondemand/
ondemandviewer.shtml?reglink=701a0000000tKKv&seri
es=buildBetterModels
1-11
Model Comparison
• A good model comparison video can be found
at:
• http://www.jmp.com/about/events/ondemand/
ondemandviewer.shtml?reglink=701a0000000tKKb&ser
ies=buildBetterModels
• We hope this help and Good Luck!