Scribd
Upload
93 views63 pages

DWDM Unit-3

This unit discusses frequent pattern mining and various association rule mining techniques. It covers mining frequent itemsets using the Apriori and FP-Growth algorithms. It also discusses mining different types of association rules, including single dimensional, multilevel, and multidimensional rules. Additionally, it covers correlation analysis and constraint-based association mining.

Uploaded by

Arun kumar Soma
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
93 views63 pages

DWDM Unit-3

This unit discusses frequent pattern mining and various association rule mining techniques. It covers mining frequent itemsets using the Apriori and FP-Growth algorithms. It also discusses mining different types of association rules, including single dimensional, multilevel, and multidimensional rules. Additionally, it covers correlation analysis and constraint-based association mining.

Uploaded by

Arun kumar Soma
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 63

Unit-3

Syllabus :
•Mining frequent Patterns,Associations and
Corelations
•Mining Methods
• Mining various kinds of Association Rules
•Correlation Analysis
•Constraint based Association Mining
INTRODUCTION
Frequent item set mining methods

1.Apriori
2.Fp-growth
3.Vertical mining algorithm-Apriori-tid
Apriori Example
Apriori Example
Drawback of Apriori
 Requires many scans
 Suitable for small datasets
Improvements of Apriori
 Partitioning
 Sampling
 Transaction reduction
 Dynamic Item set counting
 Direct Hashing and pruning
Partitioning
Sampling
Sampling (mining on a subset of the given data):
The basic idea of the sampling
approach is to pick a random sample S of the given
data D, and then search for
frequent itemsets in S instead of D.
Transaction reduction
Transaction reduction (reducing the number of transactions
scanned in future iterations):
A transaction that does not contain any frequent k-itemsets
cannot contain any frequent (k C1)-itemsets. Therefore, such
a transaction can be marked or removed from further
consideration
C1 support
Transaction reduction{bread}
3
tid Items bought
1 Bread,butter {butter} 3
2 Egg,cheese,butter
3 Bread,butter,egg {egg}  3
4 Bread,egg,cheese
5 Milk,yogurt {cheese}  2

{Milk}  1
Minimum support =2
{yogurt}  1
Dynamic Item set counting
Direct hashing and pruning
Hash-based technique (hashing itemsets into corresponding buckets): A hash-based
technique can be used to reduce the size of the candidate k-itemsets, Ck, for k > 1.
For example, when scanning each transaction in the database to generate the frequent
1-itemsets, L1, we can generate all the 2-itemsets for each transaction, hash (i.e., map)
them into the different buckets of a hash table structure, and increase the corresponding
bucket counts .
A 2-itemset with a corresponding bucket count in the hash table that is below the support threshold cannot be
frequent and thus should be removed from the candidate set. Such a hash-based technique may substantially reduce
the number of candidate k-itemsets examined (especially when k = 2).
DHP : Example
DHP : Example

The possible 2-item sets are hashed into a table.


DHP : Example
DHP : Example
DHP : Example
DHP : Example
Mining frequent item sets without
using candidate generation approach
FP-Growth
FP-Growth :steps
1) Get the support of each item
2) Delete the items from database, whose support is less than the minimum
support.
3) Sort each transaction in the database in the descending order of their
support.
4) Construct FP-tree for sorted database.
5) Find conditional pattern base.
6) Construct conditional FP-tree
7) Generate frequent item sets from conditional FP-tree.
Example
Minimum support=2
Step1
1) Get the items and support:
Item Support
I1 6
I2 7
I3 6
I4 2
I5 2
step2
2 ) Delete the items from database, whose
support is less than the minimum support:

From the previous table,no item is there in the


database, whose support is less than the
minimum support. So no item is deleted from
the database.
step3
3) Sort each transaction in the database in the
descending order of their support.
Original Database Sorted Database
Tid Sorted items
T100 I2,I1,I5
T200 I2,I4
T300 I2,I3
T400 I2,I1,I4
T500 I1,I3
T600 I2,I3
Items and their support
Item Support T700 I1,I3

I1 6 T800 I2,I1,I3,I5

I2 7 T900 I2,I1,I3

I3 6
I4 2
Step4
4)Construct FP-tree for the sorted database:
Tree construction starts from creating root node.
Root node for a FP-tree is NULL.

{ }
Step 4 : FP-tree construction
Get the second transaction from the sorted
database and insert into FP tree.
{ }
T100 = I2,I1,I5
I2 :1

I1 : 1
I5:1
Step 4 : FP-tree construction
Get the first transaction from the sorted database
and insert into FP tree.
{ }
T200 = I2,I4
I2 :1

I1 : 1
I5:1
Step 4 : FP-tree construction
Get the third transaction from the sorted database
and insert into FP tree.
{ }
T300 = I2,I3
I2 :2

I1 : 1 I4:1
I5:1
Step 4
Final FP-tree after inserting all transactions
Step5
5)Find conditional pattern base for
each item in Fp-tree.
Consider Item I5 .
Find prefix paths for I5:
{I2:1,I1:1,I5:1} {I2:1,I1:1,I3:1,I5:1}

Normalize the paths:


{I2:1,I1:1,I5:1} {I2:1,I1:1,I3:1,I5:1}
 Filter the paths:
{I2:1,I1:1} {I2:1,I1:1,I3:1}
Hence conditional pattern base for item I5
{I2:1,I1:1} {I2:1,I1:1,I3:1}
Step 5

Similarly find conditional pattern base for all the


items in the FP-tree.
Step6
6)For each conditional pattern base, construct conditional FP-tree
and generate frequent item sets from each conditional pattern
base.
Suppose consider item I3,
Conditional pattern base for I3:
{I2,I1:2}
{I2 :2}
{I1:2}
{ } { } { } { }

I2:2 I1:2
I2:4 I2:4
I1:2
I1:2
I1:2
Step6
Conditional Fp-tree ,costructed for conditional pattern base “I3”

Generating Association Rules from above conditional FP-tree:

I3
{I2 :4,I3:6} {I2,I3 :4}

{I1:4,I3:6} { I1,I3 :4}

{I2:4.I1:2,I3:6} {I2 , I1 , I3:2}


Final Frequent item sets generated
Mining various kinds of Association Rules

• Mining single dimensional association rules


• Mining multilevel association rules
• Mining multi dimensional association rules
Mining Multidimensional Association Rules
Mining Multidimensional Association Rules : Example
Mining Multidimensional Association Rules : Example
Mining Multidimensional Association Rules : Example
Mining Multidimensional Association Rules : Example
Mining Multidimensional Association Rules : Example
Co-relation Analysis
Co-relation Analysis
Co-relation Analysis
Constraint-Based Association Mining
The constraints can include the following:

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