03 Pre Processing

FP-Growth
Fatima Radi
Farah Al-Tufaili
University of Kufa – facility of computer science and math
computer science department 2015
First let’s look back to The Apriori Algorithm
 Apriori is an algorithm for frequent item set mining and

association rule learning over transactional databases.
 It proceeds by identifying the frequent individual items in
the database and extending them to larger and larger item
sets as long as those item sets appear sufficiently often in
the database.
 The frequent item sets determined by Apriori can be used
to determine association rules which highlight general
trends in the database: this has applications in domains
such as market basket analysis.
The Apriori Algorithm — Example
Database D itemset sup.
L1 itemset sup.
TID Items C1 {1} 2 {1} 2
100 134 {2} 3 {2} 3
200 235 Scan D {3} 3 {3} 3
300 1235 {4} 1 {5} 3
400 25 {5} 3
C2 itemset sup C2 itemset
L2 itemset sup {1 2} 1 Scan D {1 2}
{1 3} 2 {1 3} 2 {1 3}
{2 3} 2 {1 5} 1 {1 5}
{2 3} 2 {2 3}
{2 5} 3
{2 5} 3 {2 5}
{3 5} 2
{3 5} 2 {3 5}
C3 itemset Scan D L3 itemset sup
{2 3 5} {2 3 5} 2
Is Apriori fast enough?
Basics of Apriori algorithm
Use frequent (k-1)-itemsets to generate k-itemsets

candidates
Scan the databases to determine frequent k-itemsets
 It is costly to handle a huge number of candidate sets
 If there are 104 frequent 1-itemsts, the Apriori algorithm will

need to generate more than 107 2-itemsets and test their
frequencies
Solving Apriori Algorithm Problems?
Another Method for Frequent Itemset
Generation
 ECLAT
 FP- Growth
 AprioriDP
 Context Based Association Rule Mining Algorithm
 Node-set-based algorithms
 GUHA procedure ASSOC
 OPUS search
What Is FP-Growth
 An efficient and scalable method to complete set of frequent

patterns.
 It allow frequent item set discovery without candidate item set
generation.
 Two step approach:
1. Build a compact data structure called the FP-Tree.
2. Extracts frequent item set directly from the FP-Tree.
FP tree example (How to identify frequent
patterns using FP tree algorithm
Suppose we have the following DataBase

Step 1 - Calculate Minimum support
 First should calculate the minimum support count. Question says

minimum support should be 30%. It calculate as follows:
Minimum support count(30/100 * 8) = 2.4
As a result, 2.4 appears but to empower the easy calculation it can be

rounded to to the ceiling value. Now,
Minimum support count is ceiling(30/100 * 8) = 3 - See more at:

http://hareenlaks.blogspot.com/2011/06/fp-tree-example-how-to-
identify.html#sthash.4yIugLAd.dpuf
Step 2 - Find frequency of occurrence
 Now time to find the frequency of occurrence of each item in the

Database table. For example, item A occurs in row 1,row 2,row 3,row 4 and
row 7. Totally 5 times occurs in the Database table. You can see the
counted frequency of occurrence of each item in Table 2
TID frequency
A 5
B 6
C 3
D 6
E 4
Table 1 - Snapshot of the Database Table2 -Frequency of Occurrence

Step 3 - Prioritize the items
 In Table 2 you can see the numbers written in Red pen. Those are the
priority of each item according to it's frequency of occurrence. Item B got
the highest priority (1) due to it's highest number of occurrences. At the
same time you have opportunity to drop the items which not fulfill the
minimum support requirement. For instance, if Database contain F which
has frequency 1, then you can drop it.
TID frequency priority

A 5 3
B 6 1
C 3 5
D 6 2
E 4 4
Table2 -Frequency of Occurrence
Step 4 -Order the items according to priority
 As you see in the Table 3 new column added to the Table 1. In the
Ordered Items column all the items are queued according to it's priority,
which mentioned in the Red ink in Table 2. For example, in the case of
ordering row 1, the highest priority item is B and after that D, A and E
respectively.
Table 3 - New version of the Table 1

Step 5 -Order the items according to priority
 As a result of previous steps we got a ordered items

table (Table 3). Now it's time to draw the FP tree. We
will mention it row by row
 Row 1:
Note that all FP trees have 'null' node as the root
node. So draw the root node first and attach the
items of the row 1 one by one respectively. (See the
Figure 1) And write their occurrences in front of it.
Figure 1- FP tree for Row 1

 Row 2:
Then update the above tree (Figure 1) by entering
the items of row 2. The items of row 2 are B,D,A,E,C.
Then without creating another branch you can go
through the previous branch up to E and then you
have to create new node after that for C. This case
same as a scenario of traveling through a road to
visit the towns of the country. You should go
through the same road to achieve another town
near to the particular town.
When you going through the branch second time
you should erase one and write two for indicating
the two times you visit to that node. If you visit
through three times then write three after erase
Figure 2- FP tree for Row 1,2
two
 Row 3:
In row 3 you have to visit B,A,E and C
respectively. So you may think you
can follow the same branch again by
replacing the values of B,A,E and C .
But you can't do that you have
opportunity to come through the B.
But can't connect B to existing A
overtaking D. As a result you should
draw another A and connect it to B
and then connect new E to that A and
new C to new E.
Figure 3 - After adding third row
 Row 4:
Then row 4 contain B,D,A. Now we can just rename
the frequency of occurrences in the existing branch.
As B:4,D,A:3.
 Row 5:
n fifth raw have only item D. Now we have
opportunity draw new branch from 'null' node. See
Figure 4.
Figure 4- Connect D to null node

 Row 6:
B and D appears in row 6. So just change the B:4
to B:5 and D:3 to D:4.
 Row 7:
Attach two new nodes A and E to the D node
which hanging on the null node. Then mark D,A,E
as D:2,A:1 and E:1.
 Row 8 :(Ohh.. last row)

Attach new node C to B. Change the traverse
times.(B:6,C:1
Figure 5 - Final FP tree
Step 6 - Validation
 After the five steps the final FP tree as follows: Figure 5.
How we know is this correct?
Now count the frequency of occurrence of each item of the FP tree and
compare it with Table 2. If both counts equal, then it is positive point to
indicate your tree is correct.
Benefits of the FP-tree Structure
 Completeness
 Preserve complete information for frequent pattern mining
 Never break a long pattern of any transaction
 Compactness
 Reduce irrelevant info—infrequent items are gone
 Items in frequency descending order: the more frequently
occurring, the more likely to be shared
 Never be larger than the original database (not count node-links
and the count field)
 There exists examples of databases, where compression ratio
could be over 100
FP-Growth Complexity
 Therefore, each path in the tree will be at least partially

traversed the number of items existing in that tree path (the
depth of the tree path) * the number of items in the header.
 Complexity of searching through all paths is then bounded by
O(header_count2 * depth of tree)

03 Pre Processing

Uploaded by

Copyright:

Available Formats

03 Pre Processing

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

03 Pre Processing

Uploaded by

Copyright:

Available Formats

FP-Growth

 Apriori is an algorithm for frequent item set mining and

Basics of Apriori algorithm

Use frequent (k-1)-itemsets to generate k-itemsets

 If there are 104 frequent 1-itemsts, the Apriori algorithm will

 An efficient and scalable method to complete set of frequent

Suppose we have the following DataBase

 First should calculate the minimum support count. Question says

Minimum support count(30/100 * 8) = 2.4

As a result, 2.4 appears but to empower the easy calculation it can be

Minimum support count is ceiling(30/100 * 8) = 3 - See more at:

 Now time to find the frequency of occurrence of each item in the

Table 1 - Snapshot of the Database Table2 -Frequency of Occurrence

TID frequency priority

Table 3 - New version of the Table 1

 As a result of previous steps we got a ordered items

Figure 1- FP tree for Row 1

Figure 4- Connect D to null node

 Row 8 :(Ohh.. last row)

 After the five steps the final FP tree as follows: Figure 5.

How we know is this correct?

 Therefore, each path in the tree will be at least partially

You might also like