0% found this document useful (0 votes)
61 views15 pages

Association Analysis in Detail

This document outlines the steps in association analysis: 1) Create item sets from transaction data, 2) Identify frequent item sets that meet a minimum support threshold, 3) Generate association rules from the frequent item sets and prune rules that do not meet a minimum confidence threshold. It provides an example of applying these steps to analyze customer transaction data and generate the rule "if bread and milk are bought then diapers are also bought" since it has high confidence.

Uploaded by

Syed Zubair
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
61 views15 pages

Association Analysis in Detail

This document outlines the steps in association analysis: 1) Create item sets from transaction data, 2) Identify frequent item sets that meet a minimum support threshold, 3) Generate association rules from the frequent item sets and prune rules that do not meet a minimum confidence threshold. It provides an example of applying these steps to analyze customer transaction data and generate the rule "if bread and milk are bought then diapers are also bought" since it has high confidence.

Uploaded by

Syed Zubair
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 15

Association Analysis

in Detail
After this video you will be able to..

• Define the terms ‘support’ and ‘confidence’


• Describe the steps in association analysis
• Explain how association rules are formed
from item sets
Association Analysis Steps
1. Create item sets
{bread} {butter} {bread, milk} {bread, beer}

2. Identify frequent item sets


{bread} {bread, beer}

3. Generate rules
{bread, milk} => {diapers}
Analysis Association Dataset
ID Items
1 diapers, bread, milk
2 bread, diapers, beer, eggs
Item Sets
3 milk, diapers, beer, butter
4 bread, milk, diapers, beer
5 bread, milk, diapers, butter

If bread and milk


{bread, milk} => {diapers} are bought, then
{milk) => {bread} diapers are also
bought
Rules
Adapted from http://www-users.cs.umn.edu/~kumar/dmbook/index.php
Create Item Sets
ID Items 1-Item Sets
1 diaper, bread, milk Item Support Support =
2 bread, diaper, beer, eggs bread 4/5 frequency of
3 milk, diaper, beer, butter butter 2/5 item set
4 bread, milk, diaper, beer milk 4/5
5 bread, milk, diaper, butter beer 3/5
diaper 5/5
eggs 1/5
‘diaper’ occurs in all
transactions ‘eggs’ occurs only
once, in transaction 2
Create Item Sets
minimum support = 3/5
ID Items 1-Item Sets
1 diaper, bread, milk Item Support
2 bread, diaper, beer, eggs {bread} 4/5
3 milk, diaper, beer, butter {butter} 2/5
4 bread, milk, diaper, beer {milk} 4/5
5 bread, milk, diaper, butter {beer} 3/5
{diaper} 5/5
{eggs} 1/5
Remove these item
sets since they have
low support.
Create Item Sets
minimum support = 3/5
ID Items 2-Item Sets
1 diaper, bread, milk Item Support 1-item sets:
2 bread, diaper, beer, eggs {bread,milk} 3/5
{bread}, {milk}, {diaper}
3 milk, diaper, beer, butter {bread,beer} 2/5
4 bread, milk, diaper, beer {bread,diaper} 4/5
5 bread, milk, diaper, butter {milk,beer} 2/5
{milk,diaper} 4/5
{beer,diaper} 3/5

‘beer’ and ‘diaper’ occur


together 3 times, in
transactions 2, 3, & 4
Create Item Sets
minimum support = 3/5
ID Items 2-Item Sets
1 diaper, bread, milk Item Support 1-item sets:
2 bread, diaper, beer, eggs {bread,milk} 3/5
{bread}, {milk}, {diaper}
3 milk, diaper, beer, butter {bread,beer} 2/5
4 bread, milk, diaper, beer {bread,diaper} 4/5
5 bread, milk, diaper, butter {milk,beer} 2/5
{milk,diaper} 4/5
{beer,diaper} 3/5
Remove these item sets
since they have low support.
Create Item Sets
minimum support = 3/5
ID Items 3-Item Sets
1 diaper, bread, milk Item Support
2 bread, diaper, beer, eggs {bread,milk, 3/5
3 milk, diaper, beer, butter diaper}
4 bread, milk, diaper, beer
5 bread, milk, diaper, butter
Only 3-item set with
1-item sets: 2-item sets: support > minimum support
{bread}, {bread,milk},
{milk}, {bread,diaper},
{diaper} {milk,diaper},
{beer,diaper}
ID Items Frequent Item Sets
1 diaper, bread, milk
2 bread, diaper, beer, eggs 1-Item Sets
3 milk, diaper, beer, butter Item Support
4 bread, milk, diaper, beer {bread} 4/5 minimum support = 3/5
5 bread, milk, diaper, butter {milk} 4/5
{diaper} 5/5
2-Item Sets
Item Support
{bread,milk} 3/5 3-Item Sets
{bread,diaper} 4/5 Item Support

{milk,diaper} 4/5 {bread,milk, 3/5


diaper}
{beer,diaper} 3/5
Rule Rule
Antecedent XY Consequent
Terms
If X, then Y
Rule Confidence support for X & Y together
conf (X  Y) = supp (X U Y)
support for X
supp (X)

Itemset Support
supp (X) = # transactions with X
total # transactions
Rule Generation & Pruning
frequent item sets association rules

each k-item set 2k-2 rules!

frequent item sets significant rules


Use rule confidence to
constrain rule generation
Keep rule if confidence > minimum confidence
ID Items 3-Item Sets Rule
1 diaper, bread, milk Item Support
2 bread, diaper, beer, eggs {bread,milk, 3/5
Example
3 milk, diaper, beer, butter diaper}
min confidence = 0.95
4 bread, milk, diaper, beer
5 bread, milk, diaper, butter conf (X  Y) = supp (X U Y)
supp (X)
Candidate rule: {bread,milk}  {diaper}
conf = supp (bread,milk,diaper) = 3/5 = 3 = 1.0
supp (bread,milk) 3/5 3

Candidate rule: {bread,diaper}  {milk}


conf = supp (bread,diaper,milk) = 3/5 = 3 = 0.75
supp (bread,diaper) 4/5 4
Association Analysis Algorithms
• Use different methods to make efficient:
• item set creation
• rule generation efficient

• Algorithms:
Apriori FP Growth Eclat
Association Analysis Steps
• Item sets created from data
• Frequent item sets identified
using support
• Rules generated from
frequent item sets and
pruned using confidence

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy