A Novel Approach for Finding Frequent Medicine Set Using Maximal Apriori for Medical Application - Proceedings of All India Seminar on Biomedical Engineering 2012 (AISOBE-2012) - page 136

Table 1 Transactional
database D

TID List of medicines Item count

T001 SARIDON, DISPRIN,

ADVIL, CROCINE

4

T002 DISPRIN, PANADOL 2

T003 DISPRIN, NIMUSULIDE,

IBOBRUFIN

3

T004 SARIDON, DISPRIN,

PANADOL

3

T005 SARIDON, NIMUSULIDE 2

T006 DISPRIN, NIMUSULIDE 2

T007 SARIDON, NIMUSULIDE 2

T008 SARIDON, DISPRIN,

NIMUSULIDE, ADVIL

4

T009 SARIDON, DISPRIN,

NIMUSULIDE

3

Step 3: Once the maximal frequent transaction is found, then according to apriori
property, consider all its nonempty subsets are frequent.
Step 4: There are itemsets remaining which are not included in maximal frequent
itemset, but are frequent. Therefore, find all frequent 1-itemsets and
pruning the database consider only those transactions which contain fre-
quent 1-itemset element but are not included in the maximal frequent
transaction.
Step 5: If no such transaction is found, then return to step 7.
Step 6: Call SRMine(PruneDatabase) Procedure.

Output: Prune Database and All frequent itemsets

Example

Suppose Table 1 is transactional database with Transactional Identity Num-
ber(TID), List of Medicines, and number of items in each transaction. There are
nine transactions. Suppose the minimum support is two.
Scan the transactional database D for count of each candidate items. It is shown
in Table 2 .
Compare the candidate support count with minimum support count and remove
the infrequent medicine from Table 2 and the result is shown in Table 3
Remove infrequent items from each transactions, update item count, and sort
the transactions. as shown in Table 4 .