Transcript Mining Association Rules between Sets of Items in Large Databases

Mining Association Rules
between Sets of Items in Large
Databases
presented by Zhuang Wang
Outline
•
•
•
•
•
Introduction
Formal Model
Apriori Algorithm
Experiments
Summary
Introduction
•
Association rule:
- Association rules are used to discover elements that
co-occur frequently within a dataset consisting of
multiple independent selections of elements (such as
purchasing transactions), and to discover rules.
• Applications:
- Questions such as "if a customer purchases product A,
how likely is he to purchase product B?" and "What
products will a customer buy if he buys products C and
D?" are answered by association-finding algorithms.
(market basket analysis)
Formal Model
• Let I = I_1, I_2,. . ., I_n be a set of items.
Let T be a database of transactions.
Each transaction t in T is represented as a subset of I .
Let X be a subset of I.
• Support and Confidence:
By an association rule, we mean an implication of the
form X  I_k, where X is a set of some items in I, and I_k
is a single item in I that is not present in X.
support: probability that a transaction contains X and I_k.
P(X ,I_k)
confidence: conditional probability that a transaction
having X also contains I_k.
P(l_k | X)
Support and Confidence - Example
Transaction ID Items Bought
• Let minimum support
1
A,B,C
50%, and minimum
2
A,C
confidence 50%, we have
3
A,D
– A  C (50%, 66.6%)
4
B,E,F
– C  A (50%, 100%)
Apriori Algorithm
• To find subsets which are common to at least a minimum
confidence of the itemsets.
• Using a "bottom up" approach, where frequent itemsets
(the sets of items that follows minimum support) are
extended one item at a time (a step known as candidate
generation), and groups of candidates are tested against
the data.
• The algorithm terminates when no further successful
extensions are found.
• Generating from each large itemset, rules that use items
from the large itemset
Find Frequent Itemsets - Example
Database D
TID Items
100 1 3 4
200 2 3 5
300 1 2 3 5
400 2 5
itemset sup.
C1
{1}
2
{2}
3
Scan D
{3}
3
{4}
1
{5}
3
C2 itemset sup
L2 itemset sup
2
2
3
2
{1
{1
{1
{2
{2
{3
C3 itemset
{2 3 5}
Scan D
{1 3}
{2 3}
{2 5}
{3 5}
2}
3}
5}
3}
5}
5}
1
2
1
2
3
2
L1 itemset sup.
{1}
{2}
{3}
{5}
2
3
3
3
C2 itemset
{1 2}
Scan D
L3 itemset sup
{2 3 5} 2
{1
{1
{2
{2
{3
3}
5}
3}
5}
5}
Experiments
• We experimented with the rule mining algorithm using
the sales data obtained from a large retailing company.
• There are a total of 46,873 customer transactions in
this data. Each transaction contains the department
numbers from which a customer bought an item in
a visit.
• There are a total of 63 departments. The
algorithm finds if there is an association between
departments in the customer purchasing behavior.
• The following rules were found for a minimum support of
1% and minimum condence of 50%.
• [Tires]  [Automotive Services] (98.80, 5.79)
• [Auto Accessories], [Tires]  [Automotive Services]
(98.29, 1.47)
• [Auto Accessories]  [Automotive Services] (79.51,
11.81)
• [Automotive Services]  [Auto Accessories] (71.60, 11.81)
• [Home Laundry Appliances]  [Maintenance Agreement
Sales] (66.55, 1.25)
• [Children's Hardlines]  [Infants and Children's wear]
(66.15, 4.24)
• [Men's Furnishing]  [Men's Sportswear] (54.86, 5.21)
Summary
• Apriori, while historically significant, suffers from a
number of inefficiencies or trade-offs, which have
spawned other algorithms.
• Hash tables: uses a hash tree to store candidate
itemsets. This hash tree has item sets at the leaves and
at internal nodes
• Partitioning: Any itemset that is potentially frequent in
DB must be frequent in at least one of the partitions of
DB
• Sampling: mining on a subset of given data, need a
lower support threshold + a method to determine the
completeness.
Reference
• R. Agrawal, T. Imielinski, A. Swami: “Mining Associations between
Sets of Items in Massive Databases”, Proc. of the ACM SIGMOD
Int'l Conference on Management of Data, Washington D.C., May
1993, 207-216.
• http://knight.cis.temple.edu/~vasilis/Courses/CIS664/
• http://en.wikipedia.org/wiki/Apriori_algorithm