There are different methods for mining the frequent patterns in a data set. Among them I am discussing FP growth method for mining frequent itemsets which do not involve the generation of candidate frequent itemsets.
Let us solve the problem given in the book Data Mining Concepts and Techniques by Jiawei Han and Micheline Kamber
The transaction table is given below:
Scan the given transaction database and create 1-item item sets and the support count of each of the items (Minimum Support count considered is 2) as shown below:
Create the set L with all 1-item item sets and their respective support count in the descending order of support count which is given below
Consider each of the transaction in the transaction table and construct the tree. While constructing the tree care should be taken to consider the L order
Now conditional sub pattern bases are created which is shown below:
Images created by Shobhitha
Our aim is to provide information to the knowledge seekers.