Ans :
- Create Frequent Pattern Tree, or FP-tree by compressing the transaction database. Along with
preserving the information about the itemset, the tree structure also retains the association
among the itemset. - Divide the transaction database into a set of conditional databases. where each associated with
one frequent item or “pattern fragment,” and examines each database separately. - For each “pattern fragment,” examine its associated itemset only. Therefore, this approach
may substantially reduce the size of the itemset to be searched, along with examining the
“growth” of patterns.
