Data Mining, Fall 2011
ISOM494 / CSIS 494
Dr. Mary Malliaris
1 E. Pearson, # 384
312-915-7064
mmallia@luc.edu
Office Hours:
Tuesday: 4 - 6 pm
Thursday: 1 - 2 pm
Friday:
5 - 6 pm
Course Objectives:
The goal of the course is to introduce students to the cur

Jingxuan Sun
Dipali Janani
9/20/11
PASW stream
Type node
Apriori node
Rule 1:
There are 11.923% of people ordered do-it-yourself products/magazines and health/fitness products in
past 12 months via direct mail. Among these people, 97.735% of them also pur

NIGHT 2: DATA
TYPICAL DATA MINING TIME LINE
Projected:
Allotted Time
Actual:
Dreaded:
(Data Acquisition)
Needed:
Data Preparation
Data Analysis
INITIAL CHALLENGES IN DATA MINING
1.
What do I want to predict?
a
transaction
an individual
a household
a st

1. AA rule the association of a conclusion (Consequent) with a set of conditions (Antecedents) in
an Association Analysis model
2. Actionable this is one of the main goals of data mining whereby deciphering if the conclusions
drawn from the models provide

NIGHT 3
ASSOCIATION
ANALYSIS also
called:
MARKET BASKET
ANALYSIS
INTRODUCTION
Techniques were developed to analyze consumer
shopping patterns
Methods are descriptive
Want to find grouping of items that typically occur
together
Output generates rules an

CLASS 5
Decision
Trees
EXAMPLE
Divide into sets with only one kind of symbol, using as few
horizontal and vertical lines as possible
10
9
8
@
@
7
@
6
#
5
#
#
#
#
4
3
@
2
@
1
@
#
#
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
9
10
RULES
Rule
1: If y > 3
then pred

Regression Models:
Linear and Logistic,
and
Support Vector Machines
6-2
Regression
One of major analytic models
Linear regression
The standard ordinary least squares regression
Can apply stepwise regression
Logistic regression
When dependent data fi

NIGHT 6
Neural Networks
BRAINMAKER
Visit this site for many examples of problems
neural networks have been useful for.
http:/www.calsci.com/Applications.html
NEURAL NETWORKS
A neural network is a simplified model of the
way the human brain processes infor

NIGHT 4
Cluster Analysis
UNSUPERVISED LEARNING
Learning without a priori knowledge about the
classification of samples; learning without a
teacher.
Kohonen (1995), SelfOrganizing Maps
2
DEFINITION
Cluster
analysis is a set of methods for
constructing a (h

DATA MINING
NIGHT 1
Overview of Data Mining
WHAT IS DATA MINING?
Some common definitions:
Searching
for meaningful patterns in large data sets
Data-driven discovery of patterns in large volumes of data
Extraction of implicit, previously unknown and
une