The Apriori Algorithm is commonly used for frequent pattern mining, and FP-growth is an upgraded version of it (AKA Association Rule Mining). It's an analytical technique for identifying common patterns or correlations in data sets.
The first strategy is Mine Merge, which does not require a specific mining algorithm and can be applied to FP-growth without any changes. The second is an FP-growth implementation of the general concept of Common Counting.
Compared to Apriori, which searches the transactions for each iteration, this method only needs to scan the database twice.
This approach does not do item pairing, which makes it faster.
The database is kept in memory in a compressed form.
If the number of unique elements in the collection is n, the time complexity, in my opinion, should be O(n2). The depth of the tree determines the complexity of the search for pathways in the FP tree for each entry of the header table.