Pattern Discovery with SAS Enterprise Miner

Class meeting 3/21/2012, Thursday

Association Analysis (Total 82 minutes)

Lectures (20 minutes)

Contents

Duration

Notes

  Market basket analysis (AAEM61 Section 8.3)

 

Self-test:

Itemsets:

1. Alice, Bob, Cathy

2. Alice, Bob, David, Eric

3. Bob, Cathy, David, Eric

4. Cathy, Eric

Question: Regarding rule Bob -> Cathy, calculate its support, confident, and lift.

15 min

 

5 min

View Slide #80 to #89

Reading: AAEM61 p.8-60 to 8-63

 

Demonstrations & Exercises (50 minutes)

Demo#

Contents

Duration

Notes

4-1

 

 

About dataset BANK

Hands-on exercise 4-1:

1)      Define dataset BANK

2)      Explore BANK

5 min

5 min

View Slide #90 to #92

Reading: AAEM61 p.8-63 to 8-66

4-2

Set up the association analysis model

Hands-on exercise 4-2:

1)      Configurations

2)      Run the model

5 min

5 min

Reading: AAEM61 p.8-66 to 8-68

4-3

Understanding the results

Hands-on exercise 4-3: Examine the results

5 min

5 min

Reading: AAEM61 p.8-66 to 8-68

4-4

Plotting the 3-dimension chart for support, confidence and lift

Hands-on exercise 4-4: Plot charts

Note: Once the chart is done, by holding the ctrl key you can use the mouse to rotate the 3-D chart.

2 min

 

2 min

 

4-5

Sequence analysis

Association/Sequence analysis using SAS EM 4.3 (The SAS EM 5.3 is temporarily not available because of some bug. But you can try out the case using SAS EM 4.3.)

Hands-on exercise 4-5:

1)      Set up the model

2)      Examine the results

 

Deliverable:

The screenshots clustering results – must include the information showing your user ID at the bottom line of the SAS EM panel.

This is to show students’ participation in the class meeting.

Email address:

Isqs6347@gmail.com

Subject:

“ISQS6347 3/5/2012 <last name>”

Due midnight on 3/5, Monday

2 min

8 min

 

 

6 min

Reading: AAEM61 p.8-66 to 8-68

Reading: ADMT Chapter 8

 

Lectures (12 minutes)

Contents

Duration

Notes

  The contingency table and the assessment of association analysis *

 For more information about contingency table, check http://en.wikipedia.org/wiki/Contingency_table

Self-test:

Draw a contingency table for the following conditions:

  1. 12 students like basketball and hunting
  2. 12 students like basketball but not hunting
  3. 6 students don’t like basketball but go hunting
  4. 10 students like neither basketball nor hunting.

Given the antecedent of basketball preference, identify the rule regarding the two sports. What are the support, confidence and lift?

5 min

 

2 min

 

5 min

View Slide #93 to #98

 

 

Self-study (Optional, 2-4 hour)

Contents

Duration

Notes

   Theoretic issues of association analysis

This portion of the course provides you in-depth introduction of association analysis. There are two topics:

1) Itemset generation - Apriori principle

2) Association rule discovery and generation

2-4 hours

View slide #101 to #120

Reference: TSK chapter 6