Association - NYU Computer Science

Document technical information

Format ppt
Size 690.2 kB
First found May 22, 2018

Document content analysis

Category Also themed
Language
English
Type
not defined
Concepts
no text concepts found

Persons

Organizations

Places

Transcript

Data Mining:
Association
Mining Association
Rules in Large
Databases
 Association rule mining
 Mining single-dimensional Boolean association rules
from transactional databases
 Mining multilevel association rules from transactional
databases
 Mining multidimensional association rules from
transactional databases and data warehouse
 From association mining to correlation analysis
 Constraint-based association mining
 Summary
What Is Association
Mining?
Association rule mining:
Finding frequent patterns, associations, correlations, or
causal structures among sets of items or objects in
transaction databases, relational databases, and other
information repositories.
Applications:
Basket data analysis, cross-marketing, catalog design,
loss-leader analysis, clustering, classification, etc.
Examples.
Rule form: “Body ead [support, confidence]”.
buys(x, “diapers”)  buys(x, “beers”) [0.5%, 60%]
major(x, “CS”) ^ takes(x, “DB”) grade(x, “A”) [1%,
75%]
Association Rule: Basic
Concepts
 Given: (1) database of transactions, (2) each transaction is
a list of items (purchased by a customer in a visit)
 Find: all rules that correlate the presence of one set of
items with that of another set of items
E.g., 98% of people who purchase tires and auto
accessories also get automotive services done
 Applications
Maintenance Agreement (What the store should do to
boost Maintenance Agreement sales)
Home Electronics (What other products should the store
stocks up?)
Attached mailing in direct marketing
Detecting “ping-pong”ing of patients, faulty “collisions”
Rule Measures: Support
and Confidence
Customer
buys both
 Find all the rules X & Y  Z with
minimum confidence and support
support, s, probability that a
transaction contains {X  Y 
Z}
confidence, c, conditional
Customer
buys beer
probability that a transaction
having {X  Y} also contains Z
Transaction ID Items Bought Let minimum support 50%, and
minimum confidence 50%, we have
2000
A,B,C
A  C (50%, 66.6%)
1000
A,C
C  A (50%, 100%)
4000
A,D
5000
B,E,F
Customer
buys diaper
Association Rule Mining: A
Road Map
 Boolean vs. quantitative associations (Based on the types of values
handled)
buys(x, “SQLServer”) ^ buys(x, “DMBook”) buys(x, “DBMiner”)
[0.2%, 60%]
age(x, “30..39”) ^ income(x, “42..48K”) buys(x, “PC”) [1%, 75%]
 Single dimension vs. multiple dimensional associations
 Single level vs. multiple-level analysis
What brands of beers are associated with what brands of diapers?
Mining Association Rules
in Large Databases
 Association rule mining
 Mining single-dimensional Boolean association rules
from transactional databases
 Mining multilevel association rules from transactional
databases
 Mining multidimensional association rules from
transactional databases and data warehouse
 From association mining to correlation analysis
 Constraint-based association mining
 Summary
Mining Association Rules—
An Example
Transaction ID
2000
1000
4000
5000
Items Bought
A,B,C
A,C
A,D
B,E,F
For rule A  C:
Min. support 50%
Min. confidence 50%
Frequent Itemset Support
{A}
75%
{B}
50%
{C}
50%
{A,C}
50%
support = support({A C}) = 50%
confidence = support({A C})/support({A}) = 66.6%
The Apriori principle:
Any subset of a frequent itemset must be frequent
Mining Frequent
Itemsets: the Key Step
Find the frequent itemsets: the sets of items
that have minimum support
A subset of a frequent itemset must also be a
frequent itemset
i.e., if {AB} is a frequent itemset, both {A} and {B} should
be a frequent itemset
Iteratively find frequent itemsets with cardinality
from 1 to k (k-itemset)
Use the frequent itemsets to generate
association rules.
The Apriori Algorithm
Join Step: Ck is generated by joining Lk-1with itself
Prune Step: Any (k-1)-itemset that is not frequent
cannot be a subset of a frequent k-itemset
Pseudo-code:
Ck: Candidate itemset of size k
Lk : frequent itemset of size k
L1 = {frequent items};
for (k = 1; Lk !=; k++) do begin
Ck+1 = candidates generated from Lk;
for each transaction t in database do
Lk+1
increment the count of all candidates in Ck+1
that are contained in t
= candidates in Ck+1 with min_support
end
return k Lk;
The Apriori Algorithm —
Example
Database D
TID
100
200
300
400
itemset sup.
C1
{1}
2
{2}
3
Scan D
{3}
3
{4}
1
{5}
3
Items
134
235
1235
25
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}
How to Generate
Candidates?
Suppose the items in Lk-1 are listed in an order
Step 1: self-joining Lk-1
insert into Ck
select p.item1, p.item2, …, p.itemk-1, q.itemk-1
from Lk-1 p, Lk-1 q
where p.item1=q.item1, …, p.itemk-2=q.itemk-2, p.itemk-1 <
q.itemk-1
Step 2: pruning
forall itemsets c in Ck do
forall (k-1)-subsets s of c do
if (s is not in Lk-1) then delete c from Ck
How to Count Supports of
Candidates?
 Why counting supports of candidates a problem?
The total number of candidates can be very huge
 One transaction may contain many candidates
 Method:
Candidate itemsets are stored in a hash-tree
Leaf node of hash-tree contains a list of itemsets
and counts
Interior node contains a hash table
Subset function: finds all the candidates contained
in a transaction
Example of Generating
Candidates
 L3={abc, abd, acd, ace, bcd}
Self-joining: L3*L3
abcd from abc and abd
acde from acd and ace
Pruning:
acde is removed because ade is not in L3
 C4={abcd}
Methods to Improve Apriori’s
Efficiency
 Hash-based itemset counting: A k-itemset whose corresponding
hashing bucket count is below the threshold cannot be frequent
 Transaction reduction: A transaction that does not contain any
frequent k-itemset is useless in subsequent scans
 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, lower support
threshold + a method to determine the completeness
 Dynamic itemset counting: add new candidate itemsets only when
all of their subsets are estimated to be frequent
Visualization of Association Rule Using Plane Graph
Mining Association Rules
in Large Databases
 Association rule mining
 Mining single-dimensional Boolean association rules
from transactional databases
 Mining multilevel association rules from transactional
databases
 Mining multidimensional association rules from
transactional databases and data warehouse
 From association mining to correlation analysis
 Constraint-based association mining
 Summary
Multiple-Level Association
Rules
Food
 Items often form hierarchy.
 Items at the lower level are
expected to have lower
support.
 Rules regarding itemsets at
appropriate levels could be
quite useful.
 Transaction database can be
encoded based on
dimensions and levels
 We can explore shared multilevel mining
bread
milk
skim
Fraser
TID
T1
T2
T3
T4
T5
2%
wheat
white
Sunset
Items
{111, 121, 211, 221}
{111, 211, 222, 323}
{112, 122, 221, 411}
{111, 121}
{111, 122, 211, 221, 413}
Mining Multi-Level
Associations
 A top_down, progressive deepening approach:
 First find high-level strong rules:
milk  bread [20%, 60%].
 Then find their lower-level “weaker” rules:
2% milk  wheat bread [6%, 50%].
 Variations at mining multiple-level association rules.
Level-crossed association rules:
2% milk  Wonder wheat bread
Association rules with multiple, alternative hierarchies:
2% milk  Wonder bread
Multi-level Association: Uniform
Support vs. Reduced Support
 Uniform Support: the same minimum support for all levels
+ One minimum support threshold. No need to examine itemsets
containing any item whose ancestors do not have minimum
support.
– Lower level items do not occur as frequently. If support
threshold
too high  miss low level associations
too low  generate too many high level associations
 Reduced Support: reduced minimum support at lower
levels
There are 4 search strategies:
Level-by-level independent
Level-cross filtering by k-itemset
Level-cross filtering by single item
Controlled level-cross filtering by single item
Uniform Support
Multi-level mining with uniform support
Level 1
min_sup = 5%
Level 2
min_sup = 5%
Milk
[support = 10%]
2% Milk
Skim Milk
[support = 6%]
[support = 4%]
Back
Reduced Support
Multi-level mining with reduced support
Level 1
min_sup = 5%
Level 2
min_sup = 3%
Milk
[support = 10%]
2% Milk
Skim Milk
[support = 6%]
[support = 4%]
Back
Multi-level Association:
Redundancy Filtering
Some rules may be redundant due to “ancestor”
relationships between items.
Example
milk  wheat bread [support = 8%, confidence = 70%]
2% milk  wheat bread [support = 2%, confidence = 72%]
We say the first rule is an ancestor of the second
rule.
A rule is redundant if its support is close to the
“expected” value, based on the rule’s ancestor.
Multi-Level Mining:
Progressive Deepening
A top-down, progressive deepening approach:
 First mine high-level frequent items:
milk (15%), bread (10%)
 Then mine their lower-level “weaker” frequent
itemsets:
2% milk (5%), wheat bread (4%)
Different min_support threshold across multilevels lead to different algorithms:
If adopting the same min_support across multilevels
then toss t if any of t’s ancestors is infrequent.
If adopting reduced min_support at lower levels
then examine only those descendents whose ancestor’s
support is frequent/non-negligible.
Progressive Refinement of
Data Mining Quality
Why progressive refinement?
Mining operator can be expensive or cheap, fine or
rough
Trade speed with quality: step-by-step refinement.
Superset coverage property:
Preserve all the positive answers—allow a positive false
test but not a false negative test.
Two- or multi-step mining:
First apply rough/cheap operator (superset coverage)
Then apply expensive algorithm on a substantially
reduced candidate set (Koperski & Han, SSD’95).
Mining Association Rules
in Large Databases
 Association rule mining
 Mining single-dimensional Boolean association rules
from transactional databases
 Mining multilevel association rules from transactional
databases
 Mining multidimensional association rules from
transactional databases and data warehouse
 From association mining to correlation analysis
 Constraint-based association mining
 Summary
Multi-Dimensional
Association: Concepts
 Single-dimensional rules:
buys(X, “milk”)  buys(X, “bread”)
 Multi-dimensional rules:  2 dimensions or predicates
 Inter-dimension association rules (no repeated predicates)
age(X,”19-25”)  occupation(X,“student”)  buys(X,“coke”)
 hybrid-dimension association rules (repeated predicates)
age(X,”19-25”)  buys(X, “popcorn”)  buys(X, “coke”)
 Categorical Attributes
 finite number of possible values, no ordering among values
 Quantitative Attributes
 numeric, implicit ordering among values
Mining Association Rules
in Large Databases
 Association rule mining
 Mining single-dimensional Boolean association rules
from transactional databases
 Mining multilevel association rules from transactional
databases
 Mining multidimensional association rules from
transactional databases and data warehouse
 From association mining to correlation analysis
 Constraint-based association mining
 Summary
Interestingness
Measurements
Objective measures
Two popular measurements:
 support; and
 confidence
Subjective measures (Silberschatz &
Tuzhilin, KDD95)
A rule (pattern) is interesting if
it is unexpected (surprising to the user);
and/or
 actionable (the user can do something with it)
Criticism to Support and
Confidence
 Example 1: (Aggarwal & Yu, PODS98)
Among 5000 students
3000 play basketball
3750 eat cereal
2000 both play basket ball and eat cereal
 play basketball  eat cereal [40%, 66.7%] is misleading
because the overall percentage of students eating cereal is 75%
which is higher than 66.7%.
 play basketball  not eat cereal [20%, 33.3%] is far more
accurate, although with lower support and confidence
basketball not basketball sum(row)
cereal
2000
1750
3750
not cereal
1000
250
1250
sum(col.)
3000
2000
5000
Criticism to Support and
Confidence (Cont.)
 Example 2:
X and Y: positively correlated,
X and Z, negatively related
support and confidence of
X=>Z dominates
 We need a measure of dependent
or correlated events
corrA, B
P( A B)

P( A) P( B)
X 1 1 1 1 0 0 0 0
Y 1 1 0 0 0 0 0 0
Z 0 1 1 1 1 1 1 1
Rule Support Confidence
X=>Y 25%
50%
X=>Z 37.50%
75%
 P(B|A)/P(B) is also called the lift
of rule A => B
Other Interestingness
Measures: Interest
Interest (correlation, lift)
P( A  B)
P( A) P( B)
taking both P(A) and P(B) in consideration
P(A^B)=P(B)*P(A), if A and B are independent events
A and B negatively correlated, if the value is less than 1;
otherwise A and B positively correlated
X 1 1 1 1 0 0 0 0
Y 1 1 0 0 0 0 0 0
Z 0 1 1 1 1 1 1 1
Itemset
Support
Interest
X,Y
X,Z
Y,Z
25%
37.50%
12.50%
2
0.9
0.57
Mining Association Rules
in Large Databases
 Association rule mining
 Mining single-dimensional Boolean association rules
from transactional databases
 Mining multilevel association rules from transactional
databases
 Mining multidimensional association rules from
transactional databases and data warehouse
 From association mining to correlation analysis
 Constraint-based association mining
 Summary
Constraint-Based Mining
 Interactive, exploratory mining giga-bytes of data?
Could it be real? — Making good use of constraints!
 What kinds of constraints can be used in mining?
Knowledge type constraint: classification, association, etc.
Data constraint: SQL-like queries
Find product pairs sold together in Vancouver in Dec.’98.
Dimension/level constraints:
in relevance to region, price, brand, customer category.
Rule constraints
small sales (price < $10) triggers big sales (sum > $200).
Interestingness constraints:
strong rules (min_support  3%, min_confidence  60%).
Mining Association Rules
in Large Databases
 Association rule mining
 Mining single-dimensional Boolean association rules
from transactional databases
 Mining multilevel association rules from transactional
databases
 Mining multidimensional association rules from
transactional databases and data warehouse
 From association mining to correlation analysis
 Constraint-based association mining
 Summary
Summary
Association rule mining
probably the most significant contribution from the
database community in KDD
A large number of papers have been published
Many interesting issues have been explored
An interesting research direction
Association analysis in other types of data: spatial
data, multimedia data, time series data, etc.
References










R. Agarwal, C. Aggarwal, and V. V. V. Prasad. A tree projection algorithm for generation of frequent
itemsets. In Journal of Parallel and Distributed Computing (Special Issue on High Performance Data
Mining), 2000.
R. Agrawal, T. Imielinski, and A. Swami. Mining association rules between sets of items in large
databases. SIGMOD'93, 207-216, Washington, D.C.
R. Agrawal and R. Srikant. Fast algorithms for mining association rules. VLDB'94 487-499, Santiago,
Chile.
R. Agrawal and R. Srikant. Mining sequential patterns. ICDE'95, 3-14, Taipei, Taiwan.
R. J. Bayardo. Efficiently mining long patterns from databases. SIGMOD'98, 85-93, Seattle,
Washington.
S. Brin, R. Motwani, and C. Silverstein. Beyond market basket: Generalizing association rules to
correlations. SIGMOD'97, 265-276, Tucson, Arizona.
S. Brin, R. Motwani, J. D. Ullman, and S. Tsur. Dynamic itemset counting and implication rules for
market basket analysis. SIGMOD'97, 255-264, Tucson, Arizona, May 1997.
K. Beyer and R. Ramakrishnan. Bottom-up computation of sparse and iceberg cubes. SIGMOD'99, 359370, Philadelphia, PA, June 1999.
D.W. Cheung, J. Han, V. Ng, and C.Y. Wong. Maintenance of discovered association rules in large
databases: An incremental updating technique. ICDE'96, 106-114, New Orleans, LA.
M. Fang, N. Shivakumar, H. Garcia-Molina, R. Motwani, and J. D. Ullman. Computing iceberg queries
efficiently. VLDB'98, 299-310, New York, NY, Aug. 1998.
References (2)










G. Grahne, L. Lakshmanan, and X. Wang. Efficient mining of constrained correlated sets. ICDE'00, 512521, San Diego, CA, Feb. 2000.
Y. Fu and J. Han. Meta-rule-guided mining of association rules in relational databases. KDOOD'95, 3946, Singapore, Dec. 1995.
T. Fukuda, Y. Morimoto, S. Morishita, and T. Tokuyama. Data mining using two-dimensional optimized
association rules: Scheme, algorithms, and visualization. SIGMOD'96, 13-23, Montreal, Canada.
E.-H. Han, G. Karypis, and V. Kumar. Scalable parallel data mining for association rules. SIGMOD'97,
277-288, Tucson, Arizona.
J. Han, G. Dong, and Y. Yin. Efficient mining of partial periodic patterns in time series database.
ICDE'99, Sydney, Australia.
J. Han and Y. Fu. Discovery of multiple-level association rules from large databases. VLDB'95, 420-431,
Zurich, Switzerland.
J. Han, J. Pei, and Y. Yin. Mining frequent patterns without candidate generation. SIGMOD'00, 1-12,
Dallas, TX, May 2000.
T. Imielinski and H. Mannila. A database perspective on knowledge discovery. Communications of ACM,
39:58-64, 1996.
M. Kamber, J. Han, and J. Y. Chiang. Metarule-guided mining of multi-dimensional association rules
using data cubes. KDD'97, 207-210, Newport Beach, California.
M. Klemettinen, H. Mannila, P. Ronkainen, H. Toivonen, and A.I. Verkamo. Finding interesting rules
from large sets of discovered association rules. CIKM'94, 401-408, Gaithersburg, Maryland.
References (3)

F. Korn, A. Labrinidis, Y. Kotidis, and C. Faloutsos. Ratio rules: A new paradigm for fast, quantifiable
data mining. VLDB'98, 582-593, New York, NY.

B. Lent, A. Swami, and J. Widom. Clustering association rules. ICDE'97, 220-231, Birmingham,
England.

H. Lu, J. Han, and L. Feng. Stock movement and n-dimensional inter-transaction association rules.
SIGMOD Workshop on Research Issues on Data Mining and Knowledge Discovery (DMKD'98), 12:112:7, Seattle, Washington.

H. Mannila, H. Toivonen, and A. I. Verkamo. Efficient algorithms for discovering association rules.
KDD'94, 181-192, Seattle, WA, July 1994.

H. Mannila, H Toivonen, and A. I. Verkamo. Discovery of frequent episodes in event sequences. Data
Mining and Knowledge Discovery, 1:259-289, 1997.

R. Meo, G. Psaila, and S. Ceri. A new SQL-like operator for mining association rules. VLDB'96, 122133, Bombay, India.

R.J. Miller and Y. Yang. Association rules over interval data. SIGMOD'97, 452-461, Tucson, Arizona.

R. Ng, L. V. S. Lakshmanan, J. Han, and A. Pang. Exploratory mining and pruning optimizations of
constrained associations rules. SIGMOD'98, 13-24, Seattle, Washington.

N. Pasquier, Y. Bastide, R. Taouil, and L. Lakhal. Discovering frequent closed itemsets for association
rules. ICDT'99, 398-416, Jerusalem, Israel, Jan. 1999.
References (4)










J.S. Park, M.S. Chen, and P.S. Yu. An effective hash-based algorithm for mining association rules.
SIGMOD'95, 175-186, San Jose, CA, May 1995.
J. Pei, J. Han, and R. Mao. CLOSET: An Efficient Algorithm for Mining Frequent Closed Itemsets.
DMKD'00, Dallas, TX, 11-20, May 2000.
J. Pei and J. Han. Can We Push More Constraints into Frequent Pattern Mining? KDD'00. Boston,
MA. Aug. 2000.
G. Piatetsky-Shapiro. Discovery, analysis, and presentation of strong rules. In G. Piatetsky-Shapiro
and W. J. Frawley, editors, Knowledge Discovery in Databases, 229-238. AAAI/MIT Press, 1991.
B. Ozden, S. Ramaswamy, and A. Silberschatz. Cyclic association rules. ICDE'98, 412-421, Orlando,
FL.
J.S. Park, M.S. Chen, and P.S. Yu. An effective hash-based algorithm for mining association rules.
SIGMOD'95, 175-186, San Jose, CA.
S. Ramaswamy, S. Mahajan, and A. Silberschatz. On the discovery of interesting patterns in
association rules. VLDB'98, 368-379, New York, NY..
S. Sarawagi, S. Thomas, and R. Agrawal. Integrating association rule mining with relational database
systems: Alternatives and implications. SIGMOD'98, 343-354, Seattle, WA.
A. Savasere, E. Omiecinski, and S. Navathe. An efficient algorithm for mining association rules in
large databases. VLDB'95, 432-443, Zurich, Switzerland.
A. Savasere, E. Omiecinski, and S. Navathe. Mining for strong negative associations in a large
database of customer transactions. ICDE'98, 494-502, Orlando, FL, Feb. 1998.
References (5)


C. Silverstein, S. Brin, R. Motwani, and J. Ullman. Scalable techniques for mining causal
structures. VLDB'98, 594-605, New York, NY.
R. Srikant and R. Agrawal. Mining generalized association rules. VLDB'95, 407-419, Zurich,
Switzerland, Sept. 1995.

R. Srikant and R. Agrawal. Mining quantitative association rules in large relational tables.
SIGMOD'96, 1-12, Montreal, Canada.

R. Srikant, Q. Vu, and R. Agrawal. Mining association rules with item constraints. KDD'97, 67-73,
Newport Beach, California.

H. Toivonen. Sampling large databases for association rules. VLDB'96, 134-145, Bombay, India,
Sept. 1996.

D. Tsur, J. D. Ullman, S. Abitboul, C. Clifton, R. Motwani, and S. Nestorov. Query flocks: A
generalization of association-rule mining. SIGMOD'98, 1-12, Seattle, Washington.

K. Yoda, T. Fukuda, Y. Morimoto, S. Morishita, and T. Tokuyama. Computing optimized rectilinear
regions for association rules. KDD'97, 96-103, Newport Beach, CA, Aug. 1997.

M. J. Zaki, S. Parthasarathy, M. Ogihara, and W. Li. Parallel algorithm for discovery of association
rules. Data Mining and Knowledge Discovery, 1:343-374, 1997.

M. Zaki. Generating Non-Redundant Association Rules. KDD'00. Boston, MA. Aug. 2000.

O. R. Zaiane, J. Han, and H. Zhu. Mining Recurrent Items in Multimedia with Progressive
Resolution Refinement. ICDE'00, 461-470, San Diego, CA, Feb. 2000.
×

Report this document