Data Mining Techniques Sequential Patterns

1
Data Mining Techniques Sequential Patterns
2
Sequential Pattern Mining

• Progress in bar-code technology has made it
  possible for retail organizations to collect and
  store massive amounts of sales data, referred to
  as the basket data
• A record in such data typically consists of the
  transaction date and the items bought in the
  transaction
• Very often, data records also contain
  customer-id, particularly when the purchase has
  been made using a credit card or a frequent-buyer
  card
• Catalog companies also collect such data using
  the orders they receive

3
Sequential Pattern Mining

• An example of such a pattern is that customers
  typically rent Star Wars (????), then Empire
  Strikes Back (?????), and then Return of the
  Jedi (?????)
• These rentals need not be consecutive
• Customers who rent some other videos in between
  also support this sequential pattern
• Elements of a sequential pattern need not be
  simple items
• Computer Science and Programming Language,
  followed by Data Structure, followed by System
  Programs and Operating Systems is an example of
  a sequential pattern in which the elements are
  sets of items

4
Sequential Pattern Mining

• Given Transaction Time, Customer Id, Items Bought

Original Database
Answer Set
5
Definition

• The length of a sequence is the number of
  itemsets in the sequence
• A sequence of length k is called a k-sequence
• The support for an itemset i is defined as the
  fraction of customers who bought the items in i
  in a single transaction
• The itemset i and the 1-sequence ltigt have the
  same support
• An itemset with minimum support is called a large
  (frequent) itemset or litemset

6
AprioriAll Algorithm

• Each itemset in a large sequence must have
  minimum support
• Any large sequence must be a list of litemsets
• Finding all sequential patterns in five phases
• Sort Phase
• Litemset Phase
• Transformation Phase
• Sequence Phase
• Maximal Phase

7
AprioriAll AlgorithmSort Phase
Customer-Sequence Version of the Database
8
AprioriAll AlgorithmLitemset Phase
Apriori/DHP FP Growth
min_sup_count2
9
AprioriAll AlgorithmTransformation Phase
10
AprioriAll AlgorithmSequence Phase
Large 2-Sequences
Customer Sequences
Large 1-Sequences
2
Large 4-Sequences
Maximal Large Sequences
Large 3-Sequences
11
Sequence PhaseCandidate Generation
12
AprioriAll AlgorithmMaximal Phase

• The sequence lt(3) (4 5) (8)gt is contained in lt(7)
  (3 8) (9) (4 5 6) (8)gt, since (3) ? (3 8), (4 5)
  ? (4 5 6) and (8) ? (8)
• The sequence lt(3) (5)gt is not contained in lt(3
  5)gt (and vice versa)
• The former represents items 3 and 5 being bought
  one after the other
• The latter represents items 3 and 5 being bought
  together.
• In a set of sequences, a sequence s is maximal if
  s is not contained in any other sequence.

13
AprioriAll Algorithm
Answer Set

• With minimum support set to 25, i.e., a minimum
  support of 2 customers
• lt (30) (90)gt and lt(30) (40 70)gt are maximal
• lt(10 20) (30)gt which is only supported by
  customer 2 does not have minimum support
• lt(30)gt, lt(40)gt, lt(70)gt, lt(90)gt, lt(30) (40)gt,
  lt(30) (70)gt and lt(40 70)gt, though having minimum
  support, are not in the answer because they are
  not maximal.

14
Summary
15
Discussions

• AprioriAll algorithm will generate a huge set of
  candidate sequences
• If there are 1000 frequent sequences of length-1,
  the algorithm will generate 1000 1000 (1000
  999) / 2 1,499,500 candidate sequences
• Many scans of databases in mining
• Difficulties at mining long sequential patterns

16
Research Topics

• Time-Interval Sequential Patterns
• Time-Gap Sequential Patterns
• Non-redundant Sequential Patterns
• Constrained Sequential Pattern Mining
• Multi-dimensional Sequential Patterns
• Generalized Sequential Patterns
• Incremental Mining Sequential Patterns
• Data Stream Sequential Pattern Mining
• Interactive Mining Sequential Patterns

17
Exercise 6
A Sequence Database (min-sup 50)
Customer sequence
SID
lta(abc)(ac)d(cf)gt
10
lt(ad)c(bc)(ae)gt
20
lt(ef)(ab)(df)cbgt
30
lteg(af)cbcgt
40