SEQUENTIAL - PowerPoint PPT Presentation

1 / 10
About This Presentation
Title:

SEQUENTIAL

Description:

SEQUENTIAL PATTERNS & THE GSP ALGORITHM BY: JOE CASABONA INTRO What are Sequential Patterns? Why don't ARs suffice? The General Sequential Pattern Algorithm Finding ... – PowerPoint PPT presentation

Number of Views:49
Avg rating:3.0/5.0
Slides: 11
Provided by: Goo7524
Category:

less

Transcript and Presenter's Notes

Title: SEQUENTIAL


1
SEQUENTIAL PATTERNS THE GSP ALGORITHM
  • BY JOE CASABONA

2
INTRO
  • What are Sequential Patterns?
  • Why don't ARs suffice?
  • The General Sequential Pattern Algorithm
  • Finding Frequent Sets
  • Candidate Generation
  • Rule Generation

3
WHAT ARE SEQUENTIAL PATTERNS?
  • "Finding statistically relevant patterns between
    data examples where the values are delivered in a
    sequence." 3
  • Very similar to Association Rules, but sequence
    in this case matters.
  •  
  • There may be times when order is important. 

4
SEQUENTIAL PATTERN EXAMPLES
  • In Transaction Processing 
  •     Do customers usually buy a new controller or
    a game first after buying an Xbox?
  • In Text Mining
  •     Order of the words important for finding
    linguistic or
  • language patterns 1

5
OBJECTIVE
  • Given a set S of input data sequences, find all
    sequences that have a user-specified minimum
    support. This is called a 'frequent sequence' or
    sequential pattern. 1
  • We will use the Generalized Sequential Pattern
    Algorithm (GSP)

6
GSP
  • Similar to Apriori Algorithm
  •  Find individual items with minSupport
    (1-sequences)
  • Use them to find 2-sequences
  • Continue using k-sequences to find
    (k1)-sequences
  • Stop when there are no more frequent sequences.
  •  
  • Difference is in Candidate Generation 

7
GSP CANDIDATE GENERATION
  • Input Frequent Set k-1 (Fk-1)
  • Output Candidate Set Ck
  •  
  • How it works
  •  
  • Join Fk-1 with Fk-1
  •  Get rid of infrequent sequences (prune)
  • Note Order of items matter 

8
CANDIDATE EXAMPLE
  • F3 lt1, 2 4gt, lt1, 2 5gt, lt1 4, 5gt,
    lt1, 4 6gt, lt2 4, 5gt, lt2 4 6gt
  •  
  • After Join lt1, 2 4, 5gt, lt1, 2 4 6gt
  •  
  • After Prune lt1, 2 4, 5gt 
  •  
  • C4  lt1, 2 4, 5gt

9
RULE GENERATION
  • Objective not to generate rules, but it can be
    done. 
  •  
  • Sequential Rule Apply confidence to  Frequent
    Sequences
  •  
  • Label Sequential Rules Replace some elements in
    X with

10
RERERENCES
  • 1 The Book I am using
  •  Liu, Bing. Web Data Mining, Chapter 2
    Association Rules and Sequential Patterns.
    Springer, December, 2006 
  •  
  • Wikipedia
  • 2 "GSP Algorithm." http//en.wikipedia.org/wiki/
    GSP_AlgorithmJune 3, 2008
  • 3 "Sequence Mining." http//en.wikipedia.org/wi
    ki/Sequence_miningOct. 30, 2008
Write a Comment
User Comments (0)
About PowerShow.com