UNITII MiningFrequent,AssociationsandCorrelations 10

Mining Frequent, Associations and Correlations: Basic Concepts, Frequent Itemset Mining
Methods: Apriori Algorithm, Finding Frequent Itemsets by Confined Candidate Generation, FP-
Growth, Generating Association Rules from Frequent Itemsets, Improving the Efficiency of
Apriori, From Association Analysis to Correlation Analysis.

MiningFrequentPatternsMining
Frequent Patterns in Data Mining
ItemSet:
AnItemsetiscollectionorsetofitems
Examples:
{Computer, Printer, MSOffice}is3itemset
{Milk,Bread}is2itemset Similarly, Set
of K items is called k item set
Frequent patterns
Thesearepatternsthatappearfrequentlyinadataset.Patternsmaybeitemsets,orsub sequences.
Example:TransactionDatabase(Dataset)
Items
TID

T1 Bread,Coke,Milk

T2 Popcorn,Bread

T3 Bread,Egg,Milk.

T4 Egg,Bread,Coke, Milk

A set of items, such as Milk & Bread that appear together in a transaction data set (Also called
as Frequent Item set).
Frequent itemset mining leadstothediscoveryofassociationsandcorrelationsamong itemsin large
transactional (or) relational data sets.
Finding frequent patterns plays an essential role in mining associations, correlations, and many
other interesting relationships among data. Moreover, it helps in data classification, clustering,
and other data mining tasks.

Associationsandcorrelations
Association rule mining (or) Frequent item set mining finds interesting associationsand
relationships (correlations) in large transactional or relational data sets. This rule shows how
frequently an item set occurs in a transaction. A typical example is Market Based Analysis.
MarketBasedAnalysis is one of the key techniques used by large relations to show associations
between items. It allows retailers to identify relationships between the items that people buy
together frequently.
Thisprocessanalyzescustomerbuying habitsbyfindingassociationsbetweenthedifferent items that
customers place in their “shopping baskets”.

The discovery of these associations can help retailers develop marketing strategies by gaining
insight into which items are frequently purchased together by customers. For instance, if
customers are buying milk, how likelyare theytoalso buybread (and what kind ofbread)onthe
same trip to the supermarket? Thisinformation can lead toincreased salesby helping retailersdo
selective marketing and plan their shelf space.
Understanding these buying patterns can help to increase sales in several ways. If there is a pair
of items, X and Y, which are frequently bought together:

• Both X and Y can be placed on the same shelf, so that buyers of one item would be prompted to
buy the other.
• Promotionaldiscounts couldbeappliedtojust oneout ofthetwoitems.
• AdvertisementsonXcouldbetargetedatbuyerswhopurchaseY.
• XandYcouldbecombinedintoanewproduct,suchas havingYinflavours ofX.
Association rule: Ifthere is a pair of items, X and Y, which are frequently bought together then
association rule is represented as X ⇒Y.
Forexample,the informationthatcustomerswho purchasecomputersalsotendto buy
antivirussoftwareat the same time is represented as
Computer⇒Antivirus_Software
Measurestodiscoverinterestingnessofassociationrules
Associationrulesanalysisisatechniquetodiscoverhowitemsareassociatedtoeachother. There are
three measure to discover interestingness of association rules. Those are:
Support: Thesupportof anitem /item setisthe numberoftransactions inwhichtheitem / item set
appears, divided by the total number of transactions.
Formula:
Where,A,BareitemsandNisthetotalnumber oftransactions.
Example:Table-1ExampleTransactions

TID Items

T1 Bread,Coke,Milk

T2 Popcorn,Bread

T3 Bread,Egg,Milk.

T4 Egg,Bread,Coke,Milk

T5 Egg,Apple

Confidence:ThissaysthathowlikelyitemBispurchasedwhenitemAispurchased,
expressed as {A → B}. The Confidence of items (A and B) is the frequency or
numberoftransactionsin whichtheitems(AandB)appear,dividedbythefrequency or
number of transactions in which the item (A) appears.

Lift: This says that how likely item B is purchased when item A is purchased, expressed as an
associationrule {A→ B}. The lift is a measure to predict the performance ofanassociationrule
(targeting model).
Ifliftvalue is:

• greaterthan1meansthat itemBislikelytobeboughtifitemAisbought,
• lessthan1means that itemBisunlikelytobeboughtifitemAisbought,
• equalsto1meansthereisnoassociationbetweenitems(AandB).
TheLiftvalueisgreaterthan1meansthatitemMilkislikelytobeboughtifitemBreadis bought.
Example: TofindSupport,ConfidenceandLiftmeasureson thefollowingtransactionaldata set.
Table-2:ExampleTransactions

TID Items

T1 Bread,Milk

T2 Bread,Diaper,Burger,Eggs

T3 Milk,Diaper,Burger,Coke

T4 Bread,Milk,Diaper,Burger

T5 Bread,Milk,Diaper,Coke

Number oftransactions=5.
Support:
1 – ItemSet:
Support{Bread}=4/5=0.8=80%
Support{Diaper} = 4/ 5=0.8= 80%
Support{Milk} =4/ 5 =0.8= 80%
Support{Burger}= 3/5= 0.6=60%
Support{Coke}= 2/5= 0.4= 40%
Support{Eggs}= 1/5=0.2= 20%
2 – ItemSet:
Support{Bread,Milk}=3 / 5=0.6 =60%
Support{Milk,Diaper} =3 / 5 =0.6=60%
Support{Milk,Burger} = 2/5= 0.4= 40%
Support{Burger,Coke}=1 /5=0.2= 20%
Support{Milk,Eggs} =0/5 =0.0= 0%
3 – ItemSet:
Support{Bread,Milk,Diaper} =2 /5 =0.4 =40%
Support{Milk,Diaper,Burger} =2/ 5=0.4=40%
 MiningMethods
The most famous story about association rule mining is the “beer and diaper.” Researchers
discovered that customers who buy diapers also tend to buy beer. This classic example showsthat
there might be many interesting association rules hidden in our daily data.
Association rules help to predict the occurrence of one item based on the occurrences of other
items in a set of transactions.
AssociationrulesExamples
• Peoplewhobuybreadwillalsobuy milk;representedas{bread→milk}
• Peoplewhobuy milkwillalsobuyeggs;representedas{milk→eggs}
• Peoplewhobuybreadwillalsobuyjam;representedas {bread→ jam}
Association Rules discover the relationship between two or more attributes. It is mainly in the
form of- If antecedent than consequent. For example, a supermarket sees that there are 200
customersonFridayevening.Outofthe200customers,100boughtchicken,andoutofthe100
customers who bought chicken, 50 have bought Onions. Thus, the association rule would be- If
customers buy chicken then buy onion too, with a support of 50/200 = 25% and a confidence of
50/100=50%.
Association rule mining is a technique to identify interesting relations between different items.
Association rule mining has to:

• Findallthefrequentitems.
• Generateassociationrulesfromtheabovefrequent itemset.
There are many methods or algorithms to perform Association Rule Mining or Frequent Itemset
Mining, those are:

• Apriorialgorithm
• FP-Growthalgorithm
Apriorialgorithm
The Apriori algorithm is a classic and powerful tool in data mining used to discover frequent
itemsets and generate association rules. Imagine a grocery store database with customer
transactions. Apriori can help you find out which items frequently appear together, revealing
valuable insights like:

• Customersbuyingbreadoftenbuybutterandmilktoo. (Frequentitemset)
• 70%ofpeoplewhopurchasediapersalsobuybabywipes.(Associationrule)

HowApriorialgorithmworks:
• Bottom-up Approach: Starts with finding frequent single items, then combines them to find
frequent pairs, triplets, and so on.
• Apriori Property: If a smaller itemset isn't frequent, none of its larger versions can be either.
This "prunes" the search space for efficiency.
• Support and Confidence: Two key measures used to define how often an itemset appears and
how strong the association between items is.

LimitationsforApriorialgorithm
• Canbecomputationallyexpensiveforlargedatasets.
• Sensitivetominimumsupportandconfidencethresholds.

FP-Growthalgorithm
FP-Growthstands forFrequent PatternGrowth, andit'sasmartersiblingoftheApriorialgorithm for
mining frequent itemsets in data. But instead of brute force, it uses a clever strategyto avoid
generating and testing tons ofcandidate sets, making it much faster and more memory-efficient.
Here's its secret weapon:
• Frequent Pattern Tree (FP-Tree): This special data structure efficiently stores the frequent
itemsets andtheir relationships. Think of it as a compressedand organizedrepresentationof your
grocery store database.
• Pattern Fragment Growth:Insteadof buildingcandidatesets, FP-Growthfocuses on"growing"
smaller frequent patterns (fragments) by adding items at their frequent ends. This avoids the
costly generation and scanning of redundant patterns.

AdvantagesofFP-GrowthoverApriori
• Fasterforlargedatasets:Nomorecandidateexplosions, justtargetedpatterngrowth.
 • Lessmemoryrequired:ThecompactFP-Treeminimizes memoryusage.
• Moreversatile:Caneasilymineconditionalfrequentpatternswithoutbuildingnewtrees.

WhentoChooseFP-Growth
• Ifyou'redealingwithlargedatasetsandwant fasterresults.
• Ifmemorylimitationsareaconcern.
• Ifyouneedtomineconditionalfrequent patterns.
Remember: BothApriori andFP-Growthhavetheirstrengthsandweaknesses.Choosingthe right
tool depends on your specific data and needs.

Apriorialgorithm
Apriori algorithmwas the first algorithmthat was proposed for frequent itemset mining. It was
introduced by R Agarwal and R Srikant.
Nameofthealgorithmis Apriori becauseitusespriorknowledge offrequentitemset properties.
FrequentItemSet
• Frequent Itemset is an itemset whose support value is greater than a threshold value(support).
Apriorialgorithmuses frequent itemsets to generate association rules. To improve the efficiency
oflevel-wisegenerationoffrequentitemsets,animportantpropertyisusedcalled Apriori property
which helps by reducing the search space.
AprioriProperty
• Allsubsetsofa frequentitemsetmustbefrequent(Aprioriproperty).
• Ifanitemsetisinfrequent, allitssupersetswillbeinfrequent.

StepsinApriorialgorithm
Apriori algorithm is a sequence of steps to be followed to find the most frequent itemset in the
givendatabase. Aminimum support threshold is given inthe problemor it is assumed bythe
user.
ThestepsfollowedintheAprioriAlgorithmofdataminingare:
JoinStep:Thisstep generates(K+1) itemsetfromK-itemsetsbyjoining each itemwith itself.

Prune Step: This step scans the count of each item in thedatabase. If the candidate item does not
meet minimum support, then it is regarded as infrequent and thus it is removed. This step is
performed to reduce the size of the candidate itemsets.
Theabovejoin andtheprunesteps iterativelyuntilthemostfrequentitemsetsare achieved.
AprioriAlgorithmExample
Consider the following dataset and find frequent item sets and generate association rules for
them. Assume that minimumsupport threshold (s= 50%) and minimumconfident threshold (c =
80%).
Transaction Listofitems

T1 I1,I2,I3

T2 I2,I3,I4

T3 I4,I5

T4 I1,I2,I4

T5 I1,I2,I3,I5

T6 I1,I2,I3,I4

Solution
Findingfrequentitemsets:
Supportthreshold=50%⇒0.5*6=3⇒min_sup= 3

Step-1:
(i) Createatable containing support count ofeachitempresent indataset –Called C1 (candidate set).

Item Count

I1 4

I2 5

I3 4

I4 4

I5 2
(ii) Prune Step:Compare candidate set item’s support count with minimum support count. The
above table shows that I5 itemdoes not meet min_sup = 3, thus it is removed, only I1, I2, I3, I4
meet min_sup count.
ThisgivesusthefollowingitemsetL1.

Item Count

I1 4

Item Count
I2 5

I3 4

I4 4

Step-2:
(i) Join step: Generate candidate set C2 (2-itemset) using L1.And find out the occurrences of2-
itemset from the given dataset.

Item Count

I1,I2 4

I1,I3 3

I1,I4 2

I2,I3 4

I2,I4 3

I3,I4 2
(ii) Prune Step: Compare candidate set item’s support count with minimum support count. The
above table shows that itemsets {I1, I4} and{I3, I4} does not meet min_sup = 3, thus those are
removed.
ThisgivesusthefollowingitemsetL2.

Item Count

I1,I2 4

I1,I3 3

I2,I3 4

I2,I4 3
Step-3:
(i) Join step: Generate candidate set C3 (3-itemset) using L2.And find out the occurrences
of 3-itemset from the given dataset.
 Item Count

I1,I2,I3 3

I1,I2,I4 2

I1,I3,I4 1

I2,I3,I4 2

(ii) Prune Step:Compare candidate set item’s support count with minimum support count. The
above table showsthat itemset {I1, I2, I4},{I1,I3, I4}and{I2, I3, I4}doesnot meet min_sup = 3,
thus those are removed. Only the itemset {I1, I2, I3} meet min_sup count.
GenerateAssociationRules:
Thus, we have discovered allthe frequent item-sets. Now we need to generate strong association
rules (satisfies the minimum confidence threshold) from frequent item sets. For that we need to
calculate confidence of each rule.
ThegivenConfidencethresholdis80%.
Theallpossibleassociationrulesfromthefrequentitemset{I1,I2,I3}are:

Thisshowsthat the associationrule {I1, I3} ⇒{I2}isstrong ifminimumconfidence threshold is

80%.
Exercise1:AprioriAlgorithm
TID Items

T1 I1,I2,I5

T2 I2,I4

T3 I2,I3

T4 I1,I2,I4

T5 I1,I3

T6 I2,I3

T7 I1,I3

T8 I1,I2,I3,I5

T9 I1,I2,I3

Consider the above dataset and find frequentitem sets and generate association rules for them.
Assume that Minimum support count is 2 and minimum confident threshold (c = 50%).

Exercise2:AprioriAlgorithm
TID Items

T1 {milk,bread}

T2 {bread,sugar}

T3 {bread,butter}

T4 {milk,bread,sugar}

T5 {milk,bread,butter}

T6 {milk,bread,butter}

T7 {milk,sugar}

T8 {milk,sugar}

T9 {sugar,butter}

T10 {milk,sugar,butter}

T11 {milk,bread,butter}

Consider the above dataset and find frequentitem sets and generate association rules for them.
Assume that Minimum support count is 3 and minimum confident threshold (c = 60%).
Association Rule Mining:
Association rule mining is a popular and well researched method for discovering interesting
relations between variables in large databases.
It is intended to identify strong rules discovered in databases using different measures of
interestingness.
Based on the concept of strong rules, Rakesh Agrawaletal.introducedassociation rules.
ProblemDefinition:
Theproblem ofassociation ruleminingis definedas:

Let beaset of binaryattributescalleditems.

Let beasetoftransactionscalledthedatabase.
Eachtransactionin hasauniquetransactionID andcontainsasubsetofthe itemsin .
Aruleisdefinedasanimplicationoftheform where
and .
Thesetsofitems(forshortitemsets) and arecalledantecedent(left-hand-sideorLHS)and
consequent(right-hand-sideorRHS)oftherule respectively.
Example:
To illustrate the concepts, we use a small example from the supermarket domain. The set ofitems
is and a small database containing the items (1 codes
presence and 0 absence of an item in a transaction) is shown in the table.

An example rule for the supermarket could be meaning that if

butter and bread are bought, customers also buy milk.
 Exampledatabasewith 4items and5 transactions

TransactionID milk bread butter beer

1 1 1 0 0
2 0 0 1 0
3 0 0 0 1
4 1 1 1 0
5 0 1 0 0

Important concepts of Association Rule Mining:

The support of an itemset is defined as the proportion of transactions in the

data set which contain the itemset. In the example database, the itemset

hasasupportof sinceitoccursin20%ofall transactions

(1 out of 5 transactions).

The confidenceof a rule is defined

.
For example, the rule has a confidence of

in the database, which means that for 100% of the transactions

containing butter and bread the rule is correct (100% of the times a customer buys butter
andbread,milkisboughtaswell).Confidencecanbeinterpretedasanestimateofthe

probability , the probability of finding the RHS of the rule in transactionsunder

the condition that these transactions also contain the LHS.

The lift of a rule is defined as

 ortheratiooftheobservedsupporttothatexpectedifXandYwereindependent.The rule

has a lift of .

The conviction of a rule is defined as

The rule has aconvictionof ,

and can be interpreted as the ratio of the expected frequency that X occurs without Y(that
is to say, the frequency that the rule makes an incorrect prediction) if X and Y were
independent divided by the observed frequency of incorrect predictions.
EfficientFrequentItemsetMiningMethods:
Finding Frequent Itemsets by Confined Candidate Generation:
The Apriori Algorithm

Apriori is a seminal algorithm proposed byR. Agrawal and R. Srikant in 1994 formining
frequent itemsets for Boolean association rules.
Thenameofthealgorithm is based on thefact that thealgorithm uses prior knowledge of
frequent itemset properties.
Apriori employs an iterative approach known as a level-wise search, where k-itemsets are
used to explore (k+1)-itemsets.
First, the set of frequent 1-itemsets is found by scanning the database to accumulate the
count for each item, and collecting those items that satisfy minimum support. The
resulting set is denoted L1.Next, L1 is used to find L2, the set of frequent 2-itemsets,
which is used to find L3, and so on, until no more frequent k-itemsets can be found.
The finding of each Lkrequires one full scan of the database.
Atwo-step process isfollowed in Aprioriconsistingof joinandprune action.
Example:

TID Listof itemIDs

T100 I1,I2,I5
T200 I2,I4
T300 I2,I3
T400 I1,I2,I4
T500 I1,I3
T600 I2,I3
T700 I1,I3
T800 I1,I2,I3,I5
T900 I1,I2,I3

Thereareninetransactionsinthisdatabase,thatis,|D| =9.
Steps:
1. In the first iteration of the algorithm, each item is a member of the set of candidate1-
itemsets, C1. The algorithm simply scans all of the transactions in order to count the number of
occurrences of each item.
2. Suppose that the minimum support count required is 2, that is, min sup = 2. The set of
frequent 1-itemsets, L1, can then be determined. It consists of the candidate 1-itemsets
satisfying minimum support.In our example, all of the candidates in C1 satisfy minimum
support.
3. To discover the set of frequent 2-itemsets, L2, the algorithm uses the join L1 on L1 to
generate a candidate set of 2-itemsets, C2.No candidates are removed fromC2 during the prune
step because each subset of the candidates is also frequent.
4. Next, the transactions in D arescanned and the support count of each candidate itemsetInC2
is accumulated.
5. The set of frequent 2-itemsets, L2, is then determined, consisting of those candidate2-
itemsets in C2 having minimum support.
6. The generation of the set of candidate 3-itemsets,C3, Fromthejoin step, we first getC3 =L2x
L2= ({I1,I2,I3},{I1,I2,I5},{I1,I3,I5},{I2,I3,I4},{I2,I3,I5},{I2,I4,I5}.Based on the
Apriori property that all subsets of a frequentitemsetmust also be frequent, we can determine
that the four latter candidates cannotpossibly be frequent.

7. ThetransactionsinDarescannedinordertodetermineL3,consistingofthosecandidate 3-
itemsets in C3 having minimum support.
8. ThealgorithmusesL3xL3togenerateacandidatesetof4-itemsets,C4.
Generating Association Rules from Frequent Itemsets:
OncethefrequentitemsetsfromtransactionsinadatabaseDhavebeenfound,itis straightforward to
generate strong association rules from them.
Example:
From Association Analysis to Correlation Analysis:
A correlation measure can be used to augment the support-confidence framework for
association rules. This leads to correlation rules of the form
A=>B[support, confidence, correlation]
That is, a correlation rule is measured not only by its support and confidence but also by
the correlation between itemsets A and B. There are many different correlation measures
from which to choose. In this section, we study various correlation measures to determine
which would be good for mining large data sets.

Liftisasimplecorrelationmeasurethatisgivenasfollows.Theoccurrenceofitemset A is
independent of the occurrence of itemset B if = P(A)P(B); otherwise,
itemsets A and B are dependent and correlated as events. This definition can easily be
extended to more than two itemsets.

The lift between the occurrence of A and B can be measured by computing

If the lift(A,B)islessthan1,thentheoccurrenceofAisnegativelycorrelatedwiththe occurrence of

B.
Iftheresultingvalueisgreaterthan1,then A and B are positively correlated,meaningthat the
occurrence of one implies the occurrence of the other.
Iftheresultingvalueisequalto1,then A and B are independent and there is no correlation
between them.
 Frequent Pattern Growth Algorithm
The two primary drawbacks of the Apriori Algorithm are: 1. At each step, candidate sets have to be
built. 2. To build the candidate sets, the algorithm has to repeatedly scan the database. These two
properties inevitably make the algorithm slower. To overcome these redundant steps, a new
association-rule mining algorithm was developed named Frequent Pattern Growth Algorithm. It
overcomes the disadvantages of the Apriori algorithm by storing all the transactions in a Trie Data
Structure. Consider the following data:-

The above-given data is a hypothetical dataset of transactions with each letter representing an item.
The frequency of each individual item is computed:-

Let the minimum support be 3. A Frequent Pattern set is built which will contain all the elements
whose frequency is greater than or equal to the minimum support. These elements are stored in
descending order of their respective frequencies. After insertion of the relevant items, the set L as
follows:-
L = {K : 5, E : 4, M : 3, O : 3, Y : 3}
Now, for each transaction, the respective Ordered-Item set is built. It is done by iterating the
Frequent Pattern set and checking if the current item is contained in the transaction in question. If the
current item is contained, the item is inserted in the Ordered-Item set for the current transaction. The
following table is built for all the
transactions:

Now, all the Ordered-Item sets are inserted into a Trie Data Structure.

Inserting the set {K, E, M, O, Y}:

Here, all the items are simply linked one after the other in the order of occurrence in the set and
initialize the support count for each item as 1.

Inserting the set {K, E, O, Y}:

Till the insertion of the elements K and E, simply the support count is increased by 1. On inserting
O we can see that there is no direct link between E and O, therefore a new node for the item O is
initialized with the support count as 1 and item E is linked to this new node. On inserting Y, we
first initialize a new node for the item Y with support count as 1 and link the new node of O with
the new node of Y.

Inserting the set {K, E, M}:

Here simply the support count of each element is increased by 1.

Inserting the set {K, M, Y}:

Similar to step b), first the support count of K is increased, then new nodes for M and Y are
initialized and linked accordingly.
Inserting the set {K, E, O}:
Here simply the support counts of the respective elements are increased. Note that the support
count of the new node of item O is increased.

Now, for each item, the Conditional Pattern Base is computed which is path labels of all the paths
which lead to any node of the given item in the frequent-pattern tree. Note that the items in the
below table are arranged in the ascending order of their frequencies. Now for each item, the
Conditional Frequent Pattern Tree is built. It is done by taking the set of elements that is common
in all the paths in the Conditional Pattern Base of that item and calculating its support count by
summing the support counts of all the paths in the Conditional Pattern Base. From the Conditional
Frequent Pattern tree, the Frequent Pattern rules are generated by pairing the items of the
Conditional Frequent Pattern Tree set to the corresponding to the item as given in the below table.
For each row, two types of association rules can be inferred for example for the first row which
contains the element, the rules K -> Y and Y -> K can be inferred. To determine the valid rule, the
confidence of both the rules is calculated and the one with confidence greater than or equal to the
minimum confidence value is retained.
Improving the Efficiency of Apriori:
“How can we further improve the efficiency of Apriori-based mining?” Many variations of
the Apriori algorithm have been proposed that focus on improving the efficiency of the
original algorithm. Several of these variations are summarized as follows:

Hash-based technique (hashing itemsets into corresponding buckets): A hash-based technique

can be used to reduce the size of the candidate k-itemsets, Ck, for k > 1. For example, when
scanning each transaction in the database to generate the frequent 1-itemsets, L1, we can generate
all the 2-itemsets for each transaction, hash (i.e., map) them into the different buckets of a hash
table structure, and increase the corresponding bucket counts. A 2-itemset with a corresponding
bucket count in the hash table that is below the support threshold cannot be frequent and thus
should be removed from the candidate set. Such a hash-based technique may substantially reduce
the number of candidate k-itemsets examined.
Transaction reduction (reducing the number of transactions scanned in future iterations):
A transaction that does not contain any frequent k-itemsets cannot contain any frequent (k C1)-
itemsets. Therefore, such a transaction can be marked or removed from further consideration
because subsequent database scans for j-itemsets, where j > k, will not need to consider such a
transaction.
Partitioning (partitioning the data to find candidate itemsets): A partitioning technique can be
used that requires just two database scans to mine the frequent itemsets. It consists of two phases.
In phase I, the algorithm divides the transactions of D into n nonoverlapping partitions. If the
minimum relative support threshold for transactions in D is min sup, then the minimum support
count for a partition is min sup _ the number of transactions in that partition. For each partition, all
the local frequent itemsets (i.e., the itemsets frequent within the partition) are found. A local
frequent itemset may or may not be frequent with respect to the entire database, D. However, any
itemset that is potentially frequent with respect to D must occur as a frequent itemset in at least
one of the partitions.Therefore, all local frequent itemsets are candidate itemsets with respect to D.
The collection of frequent itemsets from all partitions forms the global candidate itemsets with
respect to D. In phase II, a second scan of D is conducted in which the actual support of each
candidate is assessed to determine the global frequent itemsets. Partition size and the number of
partitions are set so that each partition can fit into main memory and therefore be
read only once in each phase.

Sampling (mining on a subset of the given data): The basic idea of the sampling approach is to
pick a random sample S of the given data D, and then search for frequent itemsets in S instead of
D. In this way, we trade off some degree of accuracy against efficiency. The S sample size is such
that the search for frequent itemsets in S can be done in main memory, and so only one scan of the
transactions in S is required overall. Because we are searching for frequent itemsets in S rather
than in D, it is possible that we will miss some of the global frequent itemsets. To reduce this
possibility, we use a lower support threshold than minimum support to find the frequent itemsets
local to S (denoted LS). The rest of the database is then used to compute the actual frequencies of
each itemset in LS. A mechanism is used to determine whether all the global frequent itemsets are
included in LS. If LS actually contains all the frequent itemsets in D, then only one scan of D is
required. Otherwise, a second pass can be done to find the frequent itemsets that were missed in
the first pass. The sampling approach is especially beneficial when efficiency is of utmost
importance such as in computationally intensive applications that must be run frequently.
Dynamic itemset counting (adding candidate itemsets at different points during a scan): A
dynamic itemset counting technique was proposed in which the database is partitioned into blocks
marked by start points. In this variation, new candidate itemsets can be added at any start point,
unlike in Apriori, which determines new candidate itemsets only immediately before each
complete database scan. The technique uses the count-so-far as the lower bound of the actual
count. If the count-so-far passes the minimum support, the itemset is added into the frequent
itemset collection and can be used to generate longer candidates. This leads to fewer database
scans than with Apriori for finding all the frequent itemsets.