Implementasi Algoritma FP-Growth dengan Closure Table untuk Penemuan Frequent Itemset pada Keranjang Belanja

FP-growth algorithm is a data mining algorithm used to find frequent itemset in market basket data. Frequent itemset is a group of items that are often purchased together in one market basket. Analysis of frequent itemset will result in association rules. FP-growth algorithm finds frequent itemset by compressing market basket data into a tree structure called FPtree. FP-tree is then analyzed to extract frequent itemset. Market basket data is always increasing for every transaction that occurs, so the process of mining frequent itemset will re-create FP-tree from scratch repeatedly every time FP-growth algorithm executed. In order for FP-tree not to be re-created from the beginning every time FP-growth algorithm executed, FP-tree needs to be stored on storage media using format suitable for tree structure. This research stores FP-tree to table in database using closure table structure. The structure of closure table has several advantages that are suitable for storing tree structure. The results obtained from storing FP-tree to table in database using closure table structure is FP-tree that has been stored in database can be analyzed repeatedly without needing to be recreated from scratch, and only updated when market basket data increases.