You are on page 1of 11

I AT I ON R UL E M IN I NG

ASSO C
C L O SED I TE MSE T S,
T I T E M S E TS ,
FREQUEN
C IA T I ON R U L ES
AND ASSO
GROUP MEMBERS
TADIWANASHE BASIL MPUPUNI R219534J
REJOICE CHIVASA R219661Q
RUNAKORWASHE PADERA R216414P
MUNASHE TSHUMA R219669B
TRIBUTE AYLA MHAKA R2110530A
TADIWANASHE CHIWOME R216368N
TAKUDZWA CHUMACHAENDA R2110754V
PRAISE DHAURE R2111549B
GILBERT TATENDA MUSHAYI R217545X
LIVISON SISCO NDAREMA R219415Q
OUTLINE
• ASSOCIATION RULE MINING
• FREQUENT ITEMSETS
• CLOSED ITEMSETS
ASSOCIATION RULE MINING
• data mining is the process of deriving trends, patterns, and useful information from a massive amount of
data.

• The data mining process of discovering the rules that govern associations and causal objects between
sets of items is known as Association Rule Mining.

• It helps in discovering relationships between databases that seem to be independent thus developing
connections between datasets.

• association rule mining is a method for identifying frequent patterns, correlations, associations, or
causal structures in data sets found in numerous databases such as relational databases, transactional
databases, and other types of data repositories.

 An association rule has 2 parts:


1. An antecedent (if)
2. A consequent (then)
 For instance if a customer buys bread and milk, then he’s 70% likely to buy butter.
IMPORTANT DEFINITIONS
1. Support :
indicates how frequently the if/then relationship appears in the database
For example 5% support means that total 5% of transactions in the database follow the rule

2. Confidence:
confidence tells about the number of times these relationships have been found to be true. A confidence of 60%
means that 60% of the customers who purchased a milk and bread also bought butter. It denotes probability that a
transaction containing A also contains B

3. support_count (X):
Number of transactions in which X appears. If X is A union B then it is the number of transactions in which A and B
are both present

4. Itemset:
A collection of one or more items

5. Maximal itemset:
An itemset is maximal if non of its supersets are frequent.

6. Closed itemset:
An itemset is closed if none of its immediate supersets have same support count as itemset
Example
TID Items
Consider in a supermarket 1 Bread, Milk
2 Bread, Diaper, Beer, Eggs
Total transactions: 100 3 Milk, Diaper, Beer, Coke
4 Bread, Milk, Diaper, Beer
Bread: 20 5 Bread, Milk, Diaper, Coke

So, 20/100 * 100 = 20% which is the support


In those 20 transactions, butter: 9 transactions
So, 9/20* 100 = 45% which is confidence
• Single dimensional association rule
Bread => butter
Dimension : buying

• Multidimensional association rule


With 2 or more predicates or dimensions.
Occupation(I.T), Age(>22) => buys(laptop)

• Hybrid association rule


With repetative predicates or dimensions.
Time(10 am), buys(tea) => buys(sandwich)

You might also like